Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to page options Skip directly to site content

Hot Topics of the Day|PHGKB
Search PHGKB:


Hot Topics of the Day are picked by experts to capture the latest information and publications on public health genomics and precision health for various diseases and health topics. Sources include published scientific literature, reviews, blogs and popular press articles.

Sign up MyPHGKB to receive the daily hot topic email alert.

Search Archive:
Archived Hot Topics of the Day By Date
558 hot topic(s) found with the query "Machine learning"

Deep learning models across the range of skin disease.
Kaushik P Venkatesh et al. NPJ Digit Med 2024 2 (1) 32 (Posted: Feb 13, 2024 9AM)

From the abstract: "We explore the evolving landscape of diagnostic artificial intelligence (AI) in dermatology, particularly focusing on deep learning models for a wide array of skin diseases beyond skin cancer. We critically analyze the current state of AI in dermatology, its potential in enhancing diagnostic accuracy, and the challenges it faces in terms of bias, applicability, and therapeutic recommendations. "

Transparency of artificial intelligence/machine learning-enabled medical devices.
Aubrey A Shick et al. NPJ Digit Med 2024 1 (1) 21 (Posted: Jan 29, 2024 8AM)

From the article: " The United States Food and Drug Administration (FDA) is reviewing an increasing number of applications for AI/ML devices, with the number receiving FDA marketing authorization nearing seven hundred as of October 2023. AI/ML devices have unique considerations during their development and use, including those for usability, equity of access, management of performance bias, the potential for continuous learning, and stakeholder (manufacturer, patient, caregiver, healthcare provider, etc.) accountability. These considerations impact not only the responsible development and use of AI/ML devices but also the regulation of such devices"

Unsupervised machine learning to investigate trajectory patterns of COVID-19 symptoms and physical activity measured via the MyHeart Counts App and smart devices.
Varsha Gupta et al. NPJ Digit Med 2023 12 (1) 239 (Posted: Jan 02, 2024 10AM)

From the abstract: " Previous studies have associated COVID-19 symptoms severity with levels of physical activity. We therefore investigated longitudinal trajectories of COVID-19 symptoms in a cohort of healthcare workers (HCWs) with non-hospitalised COVID-19 and their real-world physical activity. 121 HCWs with a history of COVID-19 infection who had symptoms monitored through at least two research clinic visits, and via smartphone were examined. HCWs with a compatible smartphone were provided with an Apple Watch Series 4 and were asked to install the MyHeart Counts Study App to collect COVID-19 symptom data and multiple physical activity parameters"

Machine learning identifies risk factors associated with long-term sick leave following COVID-19 in Danish population
KD Jacobsen et al, Comm Med, December 20, 2023 (Posted: Dec 20, 2023 9AM)

From the abstract: "Here, in a cohort of 88,818 individuals, including 37,482 with a confirmed SARS-CoV-2 infection, the RD of long-term sick-leave is 3.3% (95% CI 3.1% to 3.6%). We observe a high degree of effect heterogeneity, with conditional RDs ranging from -3.4% to 13.7%. Age, high BMI, depression, and sex are the most important variables explaining heterogeneity. "

Organ aging signatures in the plasma proteome track health and disease.
Hamilton Se-Hwee Oh et al. Nature 2023 12 (7990) 164-172 (Posted: Dec 12, 2023 10AM)

From the abstract: "Using machine learning models, we analysed aging in 11 major organs and estimated organ age reproducibly in five independent cohorts encompassing 5,676 adults across the human lifespan. We discovered nearly 20% of the population show strongly accelerated age in one organ and 1.7% are multi-organ agers. Accelerated organ aging confers 20–50% higher mortality risk, and organ-specific diseases relate to faster aging of those organs. We find individuals with accelerated heart aging have a 250% increased heart failure risk and accelerated brain and vascular aging predict Alzheimer’s disease (AD) progression. "

Machine learning improves prediction of clinical outcomes for invasive breast cancers.
et al. Nat Med 2023 11 (Posted: Dec 01, 2023 7AM)

From the article: " A prognostic model for invasive breast cancer that is based on interpretable measurements of epithelial, stromal, and immune components outperforms histologic grading by expert pathologists. This model could improve clinical management of patients diagnosed with invasive breast cancer and address the concerns of pathologists about artificial intelligence (AI) trustworthiness by providing transparent and explainable predictions."

Natural language processing system for rapid detection and intervention of mental health crisis chat messages.
Akshay Swaminathan et al. NPJ Digit Med 2023 11 (1) 213 (Posted: Nov 22, 2023 9AM)

From the abstract: "Patients experiencing mental health crises often seek help through messaging-based platforms, but may face long wait times due to limited message triage capacity. Here we build and deploy a machine-learning-enabled system to improve response times to crisis messages in a large, national telehealth provider network. We train a two-stage natural language processing (NLP) system with key word filtering followed by logistic regression on 721 electronic medical record chat messages, of which 32% are potential crises. "

Robust airway microbiome signatures in acute respiratory failure and hospital-acquired pneumonia.
Emmanuel Montassier et al. Nat Med 2023 11 (Posted: Nov 14, 2023 9AM)

From the abstract: "Respiratory microbial dysbiosis is associated with acute respiratory distress syndrome (ARDS) and hospital-acquired pneumonia (HAP) in critically ill patients. However, we lack reproducible respiratory microbiome signatures that can increase our understanding of these conditions and potential treatments. Here, we analyze 16S rRNA sequencing data from 2,177 respiratory samples collected from 1,029 critically ill patients (21.7% with ARDS and 26.3% with HAP) and 327 healthy controls. Using machine learning models, we identified clinically informative, three- and four-factor signatures that predicted ARDS, HAP and prolonged mechanical ventilation with relatively high accuracy (area under the curve of 0.751, 0.72 and 0.727, respectively). "

Computational immunogenomic approaches to predict response to cancer immunotherapies.
Venkateswar Addala et al. Nat Rev Clin Oncol 2023 11 (Posted: Nov 03, 2023 8AM)

From the abstract: " Cancer immunogenomics is an emerging field that bridges genomics and immunology. The establishment of large-scale genomic collaborative efforts along with the development of new single-cell transcriptomic techniques and multi-omics approaches have enabled characterization of the mutational and transcriptional profiles of many cancer types and helped to identify clinically actionable alterations as well as predictive and prognostic biomarkers. Researchers have developed computational approaches and machine learning algorithms to accurately obtain clinically useful information from genomic and transcriptomic sequencing data. "

Validity of Diagnostic Support Model for Attention Deficit Hyperactivity Disorder: A Machine Learning Approach
KC Chu et al, JPM, October 25, 2023 (Posted: Oct 25, 2023 9AM)

From the abstract: "An accurate and early diagnosis of attention deficit hyperactivity disorder can improve health outcomes and prevent unnecessary medical expenses. This study developed a diagnostic support model using a machine learning approach to effectively screen individuals for attention deficit hyperactivity disorder. Three models were developed: a logistic regression model, a classification and regression tree (CART), and a neural network. The models were assessed by using a receiver operating characteristic analysis. In total, 74 participants were enrolled into the disorder group, while 21 participants were enrolled in the control group. "

Digital phenotyping could help detect autism.
Catherine Lord et al. Nat Med 2023 10 (Posted: Oct 07, 2023 8AM)

From the paper: " Researchers have developed a screening tool for autism that uses computer vision and machine learning to analyze autism-related behaviors — but greater reliability and robust validation will be needed if such tools are to be used in primary care settings."

Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease.
Shern Ping Choy et al. NPJ Digit Med 2023 9 (1) 180 (Posted: Sep 28, 2023 11AM)

From the abstract: "We searched for studies applying deep learning to skin images, excluding benign/malignant lesions. The primary outcome was accuracy of deep learning algorithms in disease diagnosis or severity assessment. We modified QUADAS-2 for quality assessment. Of 13,857 references identified, 64 were included. The most studied diseases were acne, psoriasis, eczema, rosacea, vitiligo, urticaria. Deep learning algorithms had high specificity and variable sensitivity in diagnosing these conditions. Accuracy of algorithms in diagnosing acne (median 94%, IQR 86–98; n?=?11), rosacea (94%, 90–97; n?=?4), eczema (93%, 90–99; n?=?9) and psoriasis (89%, 78–92; n?=?8) was high. "

Distinguishing features of Long COVID identified through immune profiling.
Jon Klein et al. Nature 2023 9 (Posted: Sep 27, 2023 8AM)

From the abstract: "Here, 273 individuals with or without LC were enrolled in a cross-sectional study that included multi-dimensional immune phenotyping and unbiased machine learning methods to identify biological features associated with LC. Marked differences were noted in circulating myeloid and lymphocyte populations relative to matched controls, as well as evidence of exaggerated humoral responses directed against SARS-CoV-2 among participants with LC. Further, higher antibody responses directed against non-SARS-CoV-2 viral pathogens were observed among individuals with LC, particularly Epstein-Barr virus. "

AI can help to speed up drug discovery - but only if we give it the right data.
Marissa Mock et al. Nature 2023 9 (7979) 467-470 (Posted: Sep 20, 2023 7AM)

From the paper: "Artificial-intelligence tools that enable companies to share data about drug candidates while keeping sensitive information safe can unleash the potential of machine learning and cutting-edge lab techniques, for the common good. "

Revolutionizing Cancer Research: The Impact of Artificial Intelligence in Digital Biobanking
C Frascarelli et al, J Per Med, September 2023 (Posted: Sep 18, 2023 11AM)

From the abstract: "As digital pathology and artificial intelligence (AI) have entered the precision medicine arena, biobanks are progressively transitioning from mere biorepositories to integrated computational databanks. Consequently, the application of AI and machine learning on these biobank datasets holds huge potential to profoundly impact cancer research. Methods. In this paper, we explore how AI and machine learning can respond to the digital evolution of biobanks with flexibility, solutions, and effective services. "

Considerations for addressing bias in artificial intelligence for health equity
Abramoff MD, et al, NPJ Digital Medicine, September 12, 2023 (Posted: Sep 13, 2023 0PM)

From the abstract: "Health equity is a primary goal of healthcare stakeholders: patients and their advocacy groups, clinicians, other providers and their professional societies, bioethicists, payors and value based care organizations, regulatory agencies, legislators, and creators of artificial intelligence/machine learning (AI/ML)-enabled medical devices. Lack of equitable access to diagnosis and treatment may be improved through new digital health technologies, especially AI/ML, but these may also exacerbate disparities, depending on how bias is addressed. "

Recommendations for the use of pediatric data in artificial intelligence and machine learning ACCEPT-AI
V Muralidahan et al. NPJ Digital Medicine, September 6, 2023 (Posted: Sep 06, 2023 9AM)

From the abstract: "ACCEPT-AI is a framework of recommendations for the safe inclusion of pediatric data in artificial intelligence and machine learning (AI/ML) research. It has been built on fundamental ethical principles of pediatric and AI research and incorporates age, consent, assent, communication, equity, protection of data, and technological considerations. ACCEPT-AI has been designed to guide researchers, clinicians, regulators, and policymakers and can be utilized as an independent tool, or adjunctively to existing AI/ML guidelines."

Harnessing deep learning for population genetic inference.
Xin Huang et al. Nat Rev Genet 2023 9 (Posted: Sep 05, 2023 9AM)

from the abstract: "The era of population genomics presents new challenges in analyzing the massive amounts of genomes and variants. Deep learning has demonstrated state-of-the-art performance for numerous applications involving large-scale data. Recently, deep learning approaches have gained popularity in population genetics; facilitated by the advent of massive genomic data sets, powerful computational hardware and complex deep learning architectures, they have been used to identify population structure, infer demographic history and investigate natural selection."

A machine learning classifier using 33 host immune response mRNAs accurately distinguishes viral and non-viral acute respiratory illnesses in nasal swab samples.
Rushika Pandya et al. Genome Med 2023 8 (1) 64 (Posted: Aug 30, 2023 9AM)

From the abstract: "Viral acute respiratory illnesses (viral ARIs) contribute significantly to human morbidity and mortality worldwide, but their successful treatment requires timely diagnosis of viral etiology, which is complicated by overlap in clinical presentation with the non-viral ARIs. Multiple pandemics in the twenty-first century to date have further highlighted the unmet need for effective monitoring of clinically relevant emerging viruses. Recent studies have identified conserved host response to viral infections in the blood."

Environmental and genetic predictors of human cardiovascular ageing
Nature Comm, August 21, 2023 (Posted: Aug 22, 2023 9AM)

Here we use machine learning approaches to quantify cardiovascular age from image-derived traits of vascular function, cardiac motion and myocardial fibrosis, as well as conduction traits from electrocardiograms, in 39,559 participants of UK Biobank. Cardiovascular ageing is found to be significantly associated with common or rare variants in genes regulating sarcomere homeostasis, myocardial immunomodulation, and tissue responses to biophysical stress.

At-home wearables and machine learning sensitively capture disease progression in amyotrophic lateral sclerosis
AS Gupta et al, Nature Comm. August 21, 2023 (Posted: Aug 22, 2023 9AM)

We investigate the use of wearable sensors, worn on four limbs at home during natural behavior, to quantify motor function and disease progression in 376 individuals with amyotrophic lateral sclerosis. We use an analysis approach that automatically detects and characterizes submovements from passively collected accelerometer data and produces a machine-learned severity score for each limb that is independent of clinical ratings. We show that this approach produces scores that progress faster than the gold standard Amyotrophic Lateral Sclerosis Functional Rating Scale.

Study proposes use of artificial intelligence to diagnose autism spectrum disorder
R Muniz, Medical XPress, August 2023 (Posted: Aug 18, 2023 11AM)

Much recent research proposes methods for diagnosing ASD based on machine learning but uses a single statistical parameter, ignoring brain network organization, which is the innovation featured by this study, the article notes. The analysis of fMRI data highlighted changes in certain brain regions associated with cognitive, emotional, learning and memory processes.

AI in Public Health
J Pina, ASTHO Blog, August 2023 (Posted: Aug 17, 2023 11AM)

Generative Artificial Intelligence (AI) tools have become increasingly available and accessible in recent years, empowering individuals and organizations to harness the potential of AI and machine learning. These newly available resources have sparked great curiosity within the public health community, and ASTHO members are considering the value of these tools in practice. Through ASTHO’s work in public health data modernization, and broadly in population health innovation, we’ve received many requests to address, recognize, and expound on the value and potential of AI in our field. However, as with any disruptive technology, responsible and ethical use is essential to ensure that these tools are employed in a manner that respects privacy, avoids misinformation, minimizes bias and inequities, and upholds societal well-being.

Guiding Risk Adjustment Models Toward Machine Learning Methods.
Gary E Weissman et al. JAMA 2023 8 (Posted: Aug 14, 2023 1PM)

Administrative risk adjustment models are widely used in health care and substantially affect spending, allocation of resources, and health equity. In the Medicare program, quality reporting, value-based and alternative payment models, and Medicare Advantage payments all rely on the same underlying model. However, despite advances in machine learning methods, here referring both to complex modeling approaches and to rigorous statistical practices developed by computer scientists and biostatisticians, the risk adjustment models used in Medicare have not been meaningfully updated.

Factors associated with healthy aging in Latin American populations.
Hernando Santamaria-Garcia et al. Nat Med 2023 8 (Posted: Aug 11, 2023 11AM)

We investigated the combined impact of social determinants of health (SDH), lifestyle factors, cardiometabolic factors, mental health symptoms and demographics (age, sex) on healthy aging (cognition and functional ability) across LAC countries with different levels of socioeconomic development using cross-sectional and longitudinal machine learning models (n?=?44,394 participants). Risk factors associated with social and health disparities, including SDH (ß?>?0.3), mental health (ß?>?0.6) and cardiometabolic risks (ß?>?0.22), significantly influenced healthy aging more than age and sex.

Get up to Speed on the Latest Developments in the Field! Register for the ORISE Current Issues in Genomics and Precision Public Health Online Training Event, September 7–8, 2023.
W White et al, CDC Blog Post, August 9, 2023 (Posted: Aug 09, 2023 11AM)

Advances in genomics, data science, machine learning, and artificial intelligence are transforming practice. Next generation public health and medical workforces need to understand these developments and how they can be used to benefit population health. Recognizing this challenge, Oak Ridge Institute for Science and Education (ORISE) is partnering with the Office of Genomics and Precision Public Health at the Centers for Disease Control and Prevention (CDC) to offer a free 2-day in-person training event covering the latest developments in these fields: Current Issues in Genomics and Precision Public Health – Using Genomics and Big Data to Improve Population Health and Reduce Health Inequities.

Machine learning for genetics-based classification and treatment response prediction in cancer of unknown primary.
Intae Moon et al. Nat Med 2023 8 (Posted: Aug 08, 2023 8PM)

Cancer of unknown primary (CUP) is a type of cancer that cannot be traced back to its primary site and accounts for 3–5% of all cancers. Established targeted therapies are lacking for CUP, leading to generally poor outcomes. We developed OncoNPC, a machine-learning classifier trained on targeted next-generation sequencing (NGS) data from 36,445 tumors across 22 cancer types from three institutions. Oncology NGS-based primary cancer-type classifier (OncoNPC) achieved a weighted F1 score of 0.942 for high confidence predictions (=0.9 ) on held-out tumor samples, which made up 65.2% of all the held-out samples

PhenoScore quantifies phenotypic variation for rare genetic diseases by combining facial analysis with other clinical features using a machine-learning framework.
Alexander J M Dingemans et al. Nat Genet 2023 8 (Posted: Aug 08, 2023 8PM)

We developed PhenoScore: an open-source, artificial intelligence-based phenomics framework, combining facial recognition technology with Human Phenotype Ontology data analysis to quantify phenotypic similarity. Here we show PhenoScore’s ability to recognize distinct phenotypic entities by establishing recognizable phenotypes for 37 of 40 investigated syndromes against clinical features observed in individuals with other neurodevelopmental disorders and show it is an improvement on existing approaches.

Text-based predictions of COVID-19 diagnosis from self-reported chemosensory descriptions
H Li et al, Comm Med, July 27, 2023 (Posted: Jul 28, 2023 8AM)

We developed a machine learning method based on Natural Language Processing (NLP) using Large Language Models (LLM) to predict COVID-19 diagnosis solely based on text descriptions of acute changes in chemosensation, i.e., smell, taste and chemesthesis, caused by the disease. The dataset of more than 1500 subjects was obtained from survey responses early in the COVID-19 pandemic, in Spring 2020.

Identifying healthy individuals with Alzheimer’s disease neuroimaging phenotypes in the UK Biobank
T Azevedo et al, Comm Med, July 20, 2023 (Posted: Jul 21, 2023 8AM)

We trained a Bayesian machine learning neural network model to generate a neuroimaging phenotype and AD score representing the probability of AD using structural MRI data in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) Cohort (cut-off 0.5, AUC 0.92, PPV 0.90, NPV 0.93). We go on to validate the model in an independent real-world dataset of the National Alzheimer’s Coordinating Centre (AUC 0.74, PPV 0.65, NPV 0.80) and demonstrate the correlation of the AD-score with cognitive scores in those with an AD-score above 0.5. We then apply the model to a healthy population in the UK Biobank study to identify a cohort at risk for Alzheimer’s disease.

FASDetect as a machine learning-based screening app for FASD in youth with ADHD
L Ehrig et al, NPJ Digital Medicine, July 19, 2023 (Posted: Jul 20, 2023 7AM)

Fetal alcohol-spectrum disorder (FASD) is underdiagnosed and often misdiagnosed as attention-deficit/hyperactivity disorder (ADHD). Here, we develop a screening tool for FASD in youth with ADHD symptoms. To develop the prediction model, medical record data from a University outpatient unit are assessed including 275 patients aged 0–19 years old with FASD with or without ADHD and 170 patients with ADHD without FASD aged 0–19 years old. We train 6 machine learning models based on 13 selected variables and evaluate their performance.

Genomics and Precision Public Health Issues Enrichment Event
Oak Ridge Institute for Science Education Enrichment Event, Atlanta, Georgia, September 7-8, 2023 Brand (Posted: Jul 17, 2023 8AM)

In the past decade, genomics, and precision health approaches such as big data science and machine learning have emerged as important tools for public health. Those entering the public health and medical workforces must keep pace with these evolving fields to maximize the benefit to public health. Recognizing this need, ORISE is partnering with the Office of Genomics and Precision Public Health at the Centers for Disease Control and Prevention to offer a two-day in-person enrichment event covering the latest developments in these fields.

Construction and validation of a gene expression classifier to predict immunotherapy response in primary triple-negative breast cancer
ME Mendez et al, Comm Medicine, July 10, 2023 (Posted: Jul 10, 2023 11AM)

We built machine learning models based on pre-ICI treatment gene expression profiles to construct gene expression classifiers to identify primary TNBC ICI-responder patients. This study involved 188 ICI-naïve and 721 specimens treated with ICI plus chemotherapy, including TNBC tumors, HR+/HER2- breast tumors, and other solid non-breast tumors. The 37-gene TNBC ICI predictive (TNBC-ICI) classifier performs well in predicting pathological complete response (pCR) to ICI plus chemotherapy on an independent TNBC validation cohort (AUC?=?0.86). The TNBC-ICI classifier shows better performance than other molecular signatures, including PD-1 (PDCD1) and PD-L1 (CD274) gene expression (AUC?=?0.67).

Artificial Intelligence in Molecular Medicine
B Gomes et al, NEJM, July 5, 2023 (Posted: Jul 06, 2023 8AM)

Deep learning, a powerful subset of machine learning that includes the use of deep neural networks, has had high-profile applications in image object recognition, voice recognition, autonomous driving, and virtual assistance. These approaches are now being applied in medicine to yield clinically directive medical information. In this review article, we briefly describe the methods used to generate high-dimensional molecular data and then focus on the key role that machine learning plays in the clinical application of such data.

Wearable movement-tracking data identify Parkinson's disease years before clinical diagnosis.
Ann-Kathrin Schalkamp et al. Nat Med 2023 7 (Posted: Jul 05, 2023 7AM)

Using UK Biobank, we investigated the predictive value of accelerometry in identifying prodromal Parkinson’s disease in the general population and compared this digital biomarker with models based on genetics, lifestyle, blood biochemistry or prodromal symptoms data. Machine learning models trained using accelerometry data achieved better test performance in distinguishing both clinically diagnosed Parkinson’s disease (n?=?153) (area under precision recall curve (AUPRC) 0.14?±?0.04) and prodromal Parkinson’s disease (n?=?113) up to 7?years pre-diagnosis (AUPRC 0.07?±?0.03) from the general population (n?=?33,009) compared with all other modalities tested.

Machine learning for ECG diagnosis and risk stratification of occlusion myocardial infarction.
Salah S Al-Zaiti et al. Nat Med 2023 6 (Posted: Jul 03, 2023 8AM)

Patients with occlusion myocardial infarction (OMI) and no ST-elevation on presenting electrocardiogram (ECG) are increasing in numbers. These patients have a poor prognosis and would benefit from immediate reperfusion therapy, but, currently, there are no accurate tools to identify them during initial triage. Here we report, to our knowledge, the first observational cohort study to develop machine learning models for the ECG diagnosis of OMI. Using 7,313 consecutive patients from multiple clinical sites, we derived and externally validated an intelligent model that outperformed practicing clinicians and other widely used commercial interpretation systems, substantially boosting both precision and sensitivity.

Development and Validation of a Machine Learning Prediction Model of Posttraumatic Stress Disorder After Military Deployment.
Santiago Papini et al. JAMA Netw Open 2023 6 (6) e2321273 (Posted: Jul 01, 2023 3PM)

Can risk for posttraumatic stress disorder (PTSD) be accurately predicted prior to military deployment? This diagnostic/prognostic study of 4711 US Army soldiers developed machine learning models to predict postdeployment PTSD using self-reported predictors collected before deployment. An optimal model was selected based on performance in 2 cohorts and then validated in a temporally and geographically distinct cohort, where approximately one-third of participants with the highest predicted risk accounted for an estimated 62.4% of the PTSD cases.

Co-evolution of epidemiology and artificial intelligence: challenges and opportunities.
Joohon Sung et al. Int J Epidemiol 2023 6 (Posted: Jun 24, 2023 10AM)

Artificial intelligence (AI), also often referred to as machine learning (ML) and deep learning (DL) is an automated process whereby information is extracted from a given dataset using computing techniques to create an algorithm for making predictions and/or classifications.1 The key difference between AI and classic epidemiology is that the latter builds models based on explicit assumptions about what matters and how, so that the results can be directly interpretable, whereas AI builds algorithms in essence for predictive models discovered from the data, without necessarily understanding why.

Advancing heart failure research using machine learning
MA Mohammad, The Lancet Digital Health, June 2023 (Posted: May 25, 2023 8AM)

Machine learning has demonstrated significant potential in various medical research fields and has the potential to uncover intricate associations and the ability to identify subtypes of heart failure beyond those that are currently recognised, improve risk prediction, and ultimately pave the way for personalised medicine.

De-black-boxing health AI: demonstrating reproducible machine learning computable phenotypes using the N3C-RECOVER Long COVID model in the All of Us data repository.
Emily R Pfaff et al. J Am Med Inform Assoc 2023 5 (Posted: May 24, 2023 9AM)

As part of the NIH Researching COVID to Enhance Recovery (RECOVER) Initiative, researchers with the National COVID Cohort Collaborative (N3C) devised and trained an ML-based phenotype to identify patients highly probable to have Long COVID. Supported by RECOVER, N3C and NIH's All of Us study partnered to reproduce the output of N3C's trained model in the All of Us data enclave, demonstrating model extensibility in multiple environments

Translating predictive analytics for public health practice: A case study of overdose prevention in Rhode Island.
Bennett Allen et al. Am J Epidemiol 2023 5 (Posted: May 21, 2023 8AM)

Prior applications of machine learning to population health have relied on conventional model assessment criteria, limiting the utility of models as decision supports for public health practitioners. To facilitate practitioner use of machine learning as decision support for area-level intervention, this study developed and applied four practice-based predictive model evaluation criteria (implementation capacity, preventive potential, health equity, and jurisdictional practicalities). We used a case study of overdose prevention in Rhode Island to illustrate how these criteria could inform public health practice and health equity promotion.

Machine learning for diagnosis of myocardial infarction using cardiac troponin concentrations.
Dimitrios Doudesis et al. Nat Med 2023 5 (Posted: May 12, 2023 9AM)

Although guidelines recommend fixed cardiac troponin thresholds for the diagnosis of myocardial infarction, troponin concentrations are influenced by age, sex, comorbidities and time from symptom onset. To improve diagnosis, we developed machine learning models that integrate cardiac troponin concentrations at presentation or on serial testing with clinical features and compute the Collaboration for the Diagnosis and Evaluation of Acute Coronary Syndrome (CoDE-ACS) score (0–100) that corresponds to an individual’s probability of myocardial infarction.

A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories.
Davide Placido et al. Nat Med 2023 5 (Posted: May 09, 2023 5AM)

Pancreatic cancer is an aggressive disease that typically presents late with poor outcomes, indicating a pronounced need for early detection. In this study, we applied artificial intelligence methods to clinical data from 6 million patients (24,000 pancreatic cancer cases) in Denmark (Danish National Patient Registry (DNPR)) and from 3 million patients (3,900 cases) in the United States (US Veterans Affairs (US-VA)). We trained machine learning models on the sequence of disease codes in clinical histories and tested prediction of cancer occurrence within incremental time windows (CancerRiskNet).

Perspectives on validation of clinical predictive algorithms.
Anne A H de Hond et al. NPJ Digit Med 2023 5 (1) 86 (Posted: May 08, 2023 8AM)

Machine learning has led to a surge in the development of clinical predictive algorithms. The generalizability of these algorithms often goes untested, leaving the community in the dark on their accuracy and safety when applied to a specific medical setting. We need clear objectives with respect to generalizability that align with the intended use. Journals, funding organizations, and regulatory bodies provide some guidance on generalizability requirements for clinical predictive algorithms, but a clear definition is often lacking.

Deep Learning for Epidemiologists: An introduction to neural networks.
Stylianos Serghiou et al. Am J Epidemiol 2023 5 (Posted: May 07, 2023 7AM)

Deep learning methods are increasingly being applied to problems in medicine and healthcare. However, few epidemiologists have received formal training in these methods. To bridge this gap, this article introduces the fundamentals of deep learning from an epidemiological perspective. Specifically, this article reviews core concepts in machine learning (overfitting, regularization, hyperparameters), explains several fundamental deep learning architectures (convolutional neural networks, recurrent neural networks), and summarizes training, evaluation, and deployment of models.

Deep learning model improves COPD risk prediction and gene discovery.
et al. Nat Genet 2023 4 (Posted: Apr 28, 2023 8AM)

Liability scores for chronic obstructive pulmonary disease obtained from our deep learning model improve genetic association discovery and risk prediction. We trained our model using full spirograms and noisy medical record labels obtained from self-reporting and hospital diagnostic codes, and demonstrated that the machine-learning-based phenotyping approach can be generalized to diseases that lack expert-defined annotations.

Inference of chronic obstructive pulmonary disease with deep learning on raw spirograms identifies new genetic loci and improves risk models.
Justin Cosentino et al. Nat Genet 2023 4 (Posted: Apr 20, 2023 9AM)

Here we train a deep convolutional neural network on noisy self-reported and International Classification of Diseases labels to predict COPD case–control status from high-dimensional raw spirograms and use the model’s predictions as a liability score. The machine-learning-based (ML-based) liability score accurately discriminates COPD cases and controls and predicts COPD-related hospitalization without any domain-specific knowledge.

A Predictive Model of Ischemic Heart Disease in Middle-Aged and Older Women Using Data Mining Technique
J Lim, J Per Med, April 2023 (Posted: Apr 16, 2023 6AM)

High-resolution circulating tumor DNA testing predicts survival in metastatic lung cancer clinical trials.
et al. Nat Med 2023 4 (Posted: Apr 15, 2023 8AM)

Data from circulating tumor DNA (ctDNA) testing were generated for over 1,900 samples across at least 3 time points in a phase 3 clinical trial and used to build a machine learning model to predict patient survival. The model accurately identified patients with a high risk of disease recurrence and could provide a basis for assigning therapies in phase 1/2 clinical trials.

Predicting response to enzalutamide and abiraterone in metastatic prostate cancer using whole-omics machine learning.
Anouk C de Jong et al. Nature communications 2023 4 (1) 1968 (Posted: Apr 10, 2023 7AM)

Response to androgen receptor signaling inhibitors (ARSI) varies widely in metastatic castration resistant prostate cancer (mCRPC). To improve treatment guidance, biomarkers are needed. We use whole-genomics (WGS; n?=?155) with matching whole-transcriptomics (WTS; n?=?113) from biopsies of ARSI-treated mCRPC patients for unbiased discovery of biomarkers and development of machine learning-based prediction models. Tumor mutational burden (q?<?0.001), structural variants (q?<?0.05), tandem duplications (q?<?0.05) and deletions (q?<?0.05) are enriched in poor responders, coupled with distinct transcriptomic expression profiles.

Is Medicine Ready for AI?
et al. The New England journal of medicine 2023 4 (14) e49 (Posted: Apr 06, 2023 9AM)

In this episode of “Intention to Treat,” Maia Hightower and Isaac Kohane join host Rachel Gotbaum to explore the promise and hazards of artificial-intelligence and machine-learning tools for both clinical and administrative uses in medicine.

Artificial Intelligence and Machine Learning in Clinical Medicine, 2023.
Charlotte J Haug et al. The New England journal of medicine 2023 3 (13) 1201-1208 (Posted: Apr 06, 2023 9AM)

The use of AI and machine learning in medicine has expanded beyond the reading of medical images. AI and machine-learning programs have entered medicine in many ways, including, but not limited to, helping to identify outbreaks of infectious diseases that may have an impact on public health; combining clinical, genetic, and many other laboratory outputs to identify rare and common conditions that might otherwise have escaped detection; and aiding in hospital business operations.

Stratification of Pediatric COVID-19 cases by inflammatory biomarker profiling and machine learning
D Subramanian et al, MEDRXIV, April 4, 2023 (Posted: Apr 05, 2023 5AM)

Prediction of ciprofloxacin resistance in hospitalized patients using machine learning.
Igor Mintz et al. Communications medicine 2023 3 (1) 43 (Posted: Mar 29, 2023 8AM)

Ciprofloxacin is a widely used antibiotic that has lost efficiency due to extensive resistance. We developed machine learning (ML) models that predict the probability of ciprofloxacin resistance in hospitalized patients. The ensemble models’ predictions are well-calibrated, and yield ROC-AUCs (area under the receiver operating characteristic curve) of 0.737 (95%CI 0.715–0.758) and 0.837 (95%CI 0.821–0.854) on independent test-sets for the agnostic and gnostic datasets, respectively. Shapley additive explanations analysis identifies that influential variables are related to resistance of previous infections, where patients arrived from (hospital, nursing home, etc.), and recent resistance frequencies in the hospital.

Multiomic signatures of body mass index identify heterogeneous health phenotypes and responses to a lifestyle intervention.
Kengo Watanabe et al. Nature medicine 2023 3 (Posted: Mar 22, 2023 7AM)

We report an atlas of cross-sectional and longitudinal changes in 1,111 blood analytes associated with variation in body mass index (BMI), as well as multiomic associations with host polygenic risk scores and gut microbiome composition, from a cohort of 1,277 individuals enrolled in a wellness program (Arivale). Machine learning model predictions of BMI from blood multiomics captured heterogeneous phenotypic states of host metabolism and gut microbiome composition better than BMI, which was also validated in an external cohort (TwinsUK). Moreover, longitudinal analyses identified variable BMI trajectories for different omics measures in response to a healthy lifestyle intervention.

Automated prioritization of sick newborns for whole genome sequencing using clinical natural language processing and machine learning.
Bennet Peterson et al. Genome medicine 2023 3 (1) 18 (Posted: Mar 20, 2023 7AM)

We have developed automated means to prioritize patients for rapid and whole genome sequencing (rWGS and WGS) directly from clinical notes. Our approach combines a clinical natural language processing (CNLP) workflow with a machine learning-based prioritization tool named Mendelian Phenotype Search Engine (MPSE). MPSE accurately and robustly identified NICU patients selected for WGS by clinical experts from Rady Children’s Hospital in San Diego (AUC 0.86) and the University of Utah (AUC 0.85).

An Explainable Host Genetic Severity Predictor Model for COVID-19 Patients
A Onoja et al, MEDRXIV, March 9, 2023 (Posted: Mar 10, 2023 3PM)

A host genetic severity predictor (HGSP) model was developed by combining several state-of-the-art machine learning algorithms (decision tree-based models: Random Forest and XGBoost classifiers). These models were trained using a genetic Whole Exome Sequencing (WES) dataset and clinical covariates (age and gender) formulated from a 5-fold stratified cross-validation computational strategy to randomly split the dataset to overcome model instability. Our study validated the HGSP model based on the 18 features (i.e., 16 identified candidate genetic variants and 2 covariates.

Prediction of SARS-CoV-2 transmission dynamics based on population-level cycle threshold values: A Machine learning and mechanistic modeling study
AA Khan et al, MEDRXIV, March 7, 2023 (Posted: Mar 07, 2023 6PM)

Predictive models in emergency medicine and their missing data strategies: a systematic review.
Emilien Arnaud et al. NPJ digital medicine 2023 2 (1) 28 (Posted: Feb 26, 2023 8AM)

In the field of emergency medicine (EM), the use of decision support tools based on artificial intelligence has increased markedly in recent years. In some cases, data are omitted deliberately and thus constitute “data not purposely collected” (DNPC). This accepted information bias can be managed in various ways: dropping patients with missing data, imputing with the mean, or using automatic techniques (e.g., machine learning) to handle or impute the data. Here, we systematically reviewed the methods used to handle missing data in EM research.

American Life in Realtime: a benchmark registry of health data for equitable precision health.
Ritika R Chaturvedi et al. Nature medicine 2023 2 (Posted: Feb 21, 2023 7AM)

Emerging precision health methods use large-scale person-generated health data from smartphones and wearables to better characterize and, ultimately, improve health and well-being through strategies customized to individual context and need3,4. Applying artificial intelligence and machine learning to person-generated health data allows unprecedented assessment of recursive, networked and latent associations between everyday life and health, including social, structural and environmental exposures, behaviors, biometrics, and health outcomes.

Multi-center retrospective cohort study applying deep learning to electrocardiograms to identify left heart valvular dysfunction
A Vaid et al, Comm Med, February 14, 2023 (Posted: Feb 15, 2023 7AM)

We use 617,338 ECGs paired to transthoracic echocardiograms from 123,096 patients to develop a deep learning model for detection of Mitral Regurgitation. Area Under Receiver Operating Characteristic curve (AUROC) is 0.88 (95% CI:0.88–0.89) in internal testing, and 0.81 (95% CI:0.80–0.82) in external validation. To develop a model for detection of Aortic Stenosis, we use 617,338 Echo-ECG pairs for 128,628 patients. AUROC is 0.89 (95% CI: 0.88-0.89) in internal testing, going to 0.86 (95% CI: 0.85-0.87) in external validation.

Machine-learning-aided multiplexed nanobiosensor for COVID-19 population immunity profiling
A Beisenova et al, MEDRXIV, February 8, 2023 (Posted: Feb 09, 2023 6AM)

Towards artificial intelligence-based learning health system for population-level mortality prediction using electrocardiograms
W Sun et al, NPJ Digital Medicine, February 6, 2023, (Posted: Feb 06, 2023 8AM)

We developed ECG-based machine learning models to predict risk of mortality among patients presenting to an emergency department or hospital for any reason. Using the 12-lead ECG traces and measurements from 1,605,268 ECGs from 748,773 healthcare episodes of 244,077 patients (2007–2020) in Alberta, Canada, we developed and validated ResNet-based Deep Learning (DL) and gradient boosting-based XGBoost (XGB) models to predict 30-day, 1-year, and 5-year mortality. The study demonstrates the validity of ECG-based DL mortality prediction models at the population-level that can be leveraged for prognostication at point of care.

Machine learning models for predicting severe COVID-19 outcomes in hospitals
P Wendland et al, MEDRXIV, January 30, 2023 (Posted: Jan 31, 2023 8AM)

The aim of this observational retrospective study is to improve early risk stratification of hospitalized Covid-19 patients by predicting in-hospital mortality, transfer to intensive care unit (ICU) and mechanical ventilation from electronic health record data of the first 24 hours after admission. Methods and Results Our machine learning model predicts in-hospital mortality (AUC=0.918), transfer to ICU (AUC=0.821) and the need for mechanical ventilation (AUC=0.654) from a few laboratory data of the first 24 hours after admission.

A large-scale machine learning study of sociodemographic factors contributing to COVID-19 severity
M Tumbas et al, MEDRXIV, January 29, 2023 (Posted: Jan 30, 2023 7AM)

We assemble 115 predictors for more than 3000 US counties and employ a well-defined COVID-19 severity measure derived from epidemiological dynamics modeling. We then use a number of advanced feature selection techniques from machine learning to determine which of these predictors significantly impact the disease severity. We obtain a surprisingly simple result, where only two variables are clearly and robustly selected - population density and proportion of African Americans.

Early detection of visual impairment in young children using a smartphone-based deep learning system.
Wenben Chen et al. Nature medicine 2023 1 (Posted: Jan 27, 2023 7AM)

Videos from 3,652 children (=48 months in age; 54.5% boys) were prospectively collected to develop and validate this system. For detecting visual impairment, AIS achieved an area under the receiver operating curve (AUC) of 0.940 in an internal validation set and an AUC of 0.843 in an external validation set collected in multiple ophthalmology clinics across China. In a further test of AIS for at-home implementation by untrained parents or caregivers using their smartphones, the system was able to adapt to different testing conditions and achieved an AUC of 0.859.

Making machine learning matter to clinicians: model actionability in medical decision-making
DE Ehrmann et al, NPJ Digital Medicine, January 24, 2023 (Posted: Jan 24, 2023 8AM)

Machine learning (ML) has the potential to transform patient care and outcomes. However, there are important differences between measuring the performance of ML models in silico and usefulness at the point of care. One lens to use to evaluate models during early development is actionability, which is currently undervalued. We propose a metric for actionability intended to be used before the evaluation of calibration and ultimately decision curve analysis and calculation of net benefit.

Explainable artificial intelligence for mental health through transparency and interpretability for understandability.
Joyce Dan W et al. NPJ digital medicine 2023 1 (1) 6 (Posted: Jan 21, 2023 6AM)

The literature on artificial intelligence (AI) or machine learning (ML) in mental health and psychiatry lacks consensus on what “explainability” means. In the more general XAI (eXplainable AI) literature, there has been some convergence on explainability meaning model-agnostic techniques that augment a complex model (with internal mechanics intractable for human understanding) with a simpler model argued to deliver results that humans can comprehend.

A machine-learning-derived, in silico marker for CAD identifies underdiagnosed patients.
Huynh Karina et al. Nature reviews. Cardiology 2023 1 (Posted: Jan 21, 2023 6AM)

Current risk prediction tools for coronary artery disease (CAD) do not measure disease on a continuous scale and use only a small number of variables for risk prediction, disregarding much of the data contained in electronic health records (EHRs). In a new study, a machine learning model trained using clinical data from EHRs generated a novel, in silico quantitative score for CAD that can quantify disease pathophysiology and clinical outcomes on a continuous spectrum.

Wearables and AI better predict the progression of muscular dystrophy
Nature Medicine, January 20, 2023 (Posted: Jan 21, 2023 6AM)

Clinical trials in neurological diseases often involve subjective, qualitative endpoints, such ‘by eye’ observations of movement. We developed an artificial intelligence–based method to analyze natural daily behavior data from people with Duchenne muscular dystrophy, using machine-learning algorithms to accurately predict their personal disease trajectories better than conventional clinical assessments.

Machine learning identifies long COVID patterns from electronic health records.
et al. Nature medicine 2023 1 (Posted: Jan 17, 2023 9AM)

A machine learning algorithm identifies four reproducible clinical subphenotypes of long COVID from the electronic health records of patients with post-acute sequelae of SARS-CoV-2 infection within 30–180 days of infection; these patterns have implications for the treatment and management of long COVID.

AI in the hands of imperfect users
KMK QUenet et al, NPJ Digital Medicine, December 28, 2022 (Posted: Dec 28, 2022 11AM)

As the use of artificial intelligence and machine learning (AI/ML) continues to expand in healthcare, much attention has been given to mitigating bias in algorithms to ensure they are employed fairly and transparently. Less attention has fallen to addressing potential bias among AI/ML’s human users or factors that influence user reliance. We argue for a systematic approach to identifying the existence and impacts of user biases while using AI/ML tools and call for the development of embedded interface design features.

Identifying and evaluating barriers for the implementation of machine learning in the intensive care unit
E D'hondt et al, Comm Med, December 21, 2022 (Posted: Dec 21, 2022 8AM)

Dietary metabolic signatures and cardiometabolic risk.
Shah Ravi V et al. European heart journal 2022 11 (Posted: Dec 17, 2022 9AM)

Observational studies of diet in cardiometabolic-cardiovascular disease (CM-CVD) focus on self-reported consumption of food or dietary pattern, with limited information on individual metabolic responses to dietary intake linked to CM-CVD. Here, machine learning approaches were used to identify individual metabolic patterns related to diet and relation to long-term CM-CVD in early adulthood. We found that metabolic signatures of diet are associated with long-term CM-CVD independent of lifestyle and traditional risk factors. Metabolomics improves precision to identify adverse consequences and pathways of diet-related CM-CVD.

Deep learning-based age estimation from chest X-rays indicates cardiovascular prognosis.
Ieki Hirotaka et al. Communications medicine 2022 12 (1) 159 (Posted: Dec 12, 2022 9AM)

In recent years, there has been considerable research on the use of artificial intelligence to estimate age and disease status from medical images. However, age estimation from chest X-ray (CXR) images has not been well studied and the clinical significance of estimated age has not been fully determined. To address this, we trained a deep neural network (DNN) model using more than 100,000 CXRs to estimate the patients' age solely from CXRs. We applied our DNN to CXRs of 1562 consecutive hospitalized heart failure patients, and 3586 patients admitted to the intensive care unit with cardiovascular disease.

Improving Machine Learning Diabetes Prediction Models for the Utmost Clinical Effectiveness
J Shin et al, J Per Med, November 14, 2022 (Posted: Nov 15, 2022 9AM)

An immune dysfunction score for stratification of patients with acute infection based on whole-blood gene expression.
Cano-Gamez Eddie et al. Science translational medicine 2022 11 (669) eabq4433 (Posted: Nov 14, 2022 6AM)

Predictors of severe infection could help physicians manage clinical care. Cano-Gamez et al. present an RNA-seq–based gene expression signature derived from patients with sepsis that generally captured patient prognosis with high sensitivity. Biologically, this signature corresponded to immune dysfunction. A machine learning framework based on the gene signature correctly stratified pediatric and adult patients with bacterial or viral sepsis, as well as patients with infection who did not meet sepsis criteria, including H1N1 influenza and COVID-19.

Health, socioeconomic and genetic predictors of COVID-19 vaccination uptake: a nationwide machine-learning study
T Hartonen et al, MEDRXIV, November 11, 2022 (Posted: Nov 12, 2022 6AM)

Broad-capture proteomics and machine learning for early detection of type 2 diabetes risk
Nature Medicine, November 10, 2022 (Posted: Nov 11, 2022 7AM)

Impaired glucose tolerance (IGT) is a common condition that affects glucose control after sugar consumption. Isolated IGT is undetected by screening and diagnostic strategies, leaving affected individuals at high risk of developing diabetes. Here, a machine-learning framework identifies a three-protein signature for detecting isolated IGT from a single blood sample.

Proteomic signatures for identification of impaired glucose tolerance
JC Zanini et al, Nature Medicine, November 10, 2022 (Posted: Nov 11, 2022 7AM)

We applied machine learning to the proteomic profiles of a single fasted sample from 11,546 participants of the Fenland study to test discrimination of iIGT defined using the gold-standard OGTTs. We observed significantly improved discriminative performance by adding only three proteins (RTN4R, CBPM and GHR) to the best clinical model (AUROC = 0.80 (95% confidence interval: 0.79–0.86), P?=?0.004), which we validated in an external cohort. Increased plasma levels of these candidate proteins were associated with an increased risk for future T2D in an independent cohort and were also increased in individuals genetically susceptible to impaired glucose homeostasis and T2D.

A comprehensive knowledgebase of known and predicted human genetic variants associated with COVID-19 susceptibility and severity
Me Kars et al, MEDRXIV, November 7, 2022 (Posted: Nov 08, 2022 7AM)

We collated 820 host genetic variants reported to affect COVID-19 susceptibility by means of a systematic literature search and confidence evaluation, and obtained 196 high-confidence variants. We then developed the first machine learning classifier of severe COVID-19 variants to perform a genome-wide prediction of COVID-19 severity for 82,468,698 missense variants in the human genome.

Multimodal machine learning in precision health: A scoping review
A Kline et al, NPJ Digital Medicine, November 7, 2022 (Posted: Nov 08, 2022 7AM)

Machine learning is frequently being leveraged to tackle problems in the health sector including utilization for clinical decision-support. Its use has historically been focused on single modal data. Attempts to improve prediction and mimic the multimodal nature of clinical expert decision-making has been met in the biomedical field of machine learning by fusing disparate data. This review was conducted to summarize the current studies in this field and identify topics ripe for future research.

Discovering a trans-omics biomarker signature that predisposes high risk diabetic patients to diabetic kidney disease
I W Wu et al, NPJ Digital Medicine, November 2, 2022 (Posted: Nov 03, 2022 8AM)

Diabetic kidney disease is the leading cause of end-stage kidney disease worldwide; however, the integration of high-dimensional trans-omics data to predict this diabetic complication is rare. We develop artificial intelligence (AI)-assisted models using machine learning algorithms to identify a biomarker signature that predisposes high risk patients with diabetes mellitus (DM) to diabetic kidney disease based on clinical information, untargeted metabolomics, targeted lipidomics and genome-wide single nucleotide polymorphism (SNP) datasets. This involves 618 individuals who are split into training and testing cohorts of 557 and 61 subjects, respectively.

Prediction of atrial fibrillation and stroke using machine learning models in UK Biobank.
A Papadopolou et al, MEDRXIV, October 30, 2022 (Posted: Oct 31, 2022 9AM)

Machine learning models for predicting severe COVID-19 outcomes in hospitals
P Wendland et al, MEDRXIV, October 30, 2022 (Posted: Oct 31, 2022 9AM)

The aim of this observational retrospective study is to improve early risk stratification of hospitalized Covid-19 patients by predicting in-hospital mortality, transfer to intensive care unit (ICU) and mechanical ventilation from electronic health record data of the first 24 hours after admission. Methods and Results Our machine learning model predicts in-hospital mortality (AUC=0.918), transfer to ICU (AUC=0.821) and the need for mechanical ventilation (AUC=0.654) from a few laboratory data of the first 24 hours after admission.

Spatio-temporal dynamics of three diseases caused by Aedes-borne arboviruses in Mexico
B Dong et al, Comm Med, October 28, 2022 (Posted: Oct 29, 2022 10AM)

We present an integrated analysis of Aedes-borne diseases (ABDs), the local climate, and the socio-demographic profiles of 2469 municipalities in Mexico. We used SaTScan to detect spatial clusters and utilize the Pearson correlation coefficient, Randomized Dependence Coefficient, and SHapley Additive exPlanations to analyze the influence of socio-demographic and climatic factors on the prevalence of ABDs. We also compare six machine learning techniques, including XGBoost, decision tree, Support Vector Machine with Radial Basis Function kernel, K nearest neighbors, random forest, and neural network to predict risk factors of ABDs clusters.

Algorithms and the Future of Work
J Howard, NIOSH Blog, September 2022 (Posted: Oct 28, 2022 9AM)

In the future of work, algorithms will provide many beneficial applications in occupational safety and health. While algorithm-enabled systems and devices may reduce sources of human error and enhance worker safety and health, algorithms may also introduce new sources of risks to worker wellbeing. To ensure that the benefits of algorithm-enabled systems and devices have a prominent place in the future of work, now is the time to study how to effectively manage their risks.

Artificial intelligence and machine learning in cancer imaging
DM Koh, Comm Medicine, October 27, 2022 (Posted: Oct 27, 2022 9AM)

An increasing array of tools is being developed using artificial intelligence (AI) and machine learning (ML) for cancer imaging. The development of an optimal tool requires multidisciplinary engagement to ensure that the appropriate use case is met, as well as to undertake robust development and testing prior to its adoption into healthcare systems. This multidisciplinary review highlights key developments in the field

Robust Machine Learning predicts COVID-19 Disease Severity based on Single-cell RNA-seq from multiple hospitals
A Lemsara et al, MEDRXIV, October 22, 2022 (Posted: Oct 24, 2022 10AM)

We present a computational workflow based on a Multilayer perceptron network that predicts the necessity of mechanical ventilation from PBMCs single-cell RNA-seq data. The study includes patient cohorts from Bonn, Berlin, Stanford, and three Korean medical centers. Training and model validation are performed using Berlin and Bonn samples, while testing is performed on completely unseen samples from the Stanford and Korean datasets. Our model shows a high area under the receiver operating characteristic (AUROC) curve (Korea: 1 (CI:1-1), Stanford: 0.86 (CI:0.81-0.9)), proving our models robustness.

Intelligent risk prediction in public health using wearable device data
MM Raza et al, NPJ Digital Medicine, October 13, 2022 (Posted: Oct 13, 2022 6AM)

A recent study found that machine learning can predict the risk of COVID-19 infection, by combining biometric data from wearable devices like Fitbit, with electronic symptom surveys. In doing so, they aim to increase the efficiency of test allocation when tracking disease spread in resource-limited settings. But the implications of technology that applies data from wearables stretch far beyond infection monitoring into healthcare delivery and research. The adoption and implementation of this type of technology will depend on regulation, impact on patient outcomes, and cost savings.

Interpretable machine learning prediction of all-cause mortality
Q Qiu et al, Comm Medicine, October 3, 2022 (Posted: Oct 04, 2022 8AM)

We propose the IMPACT framework that uses XAI technique to explain a state-of-the-art tree ensemble mortality prediction model. We apply IMPACT to understand all-cause mortality for 1-, 3-, 5-, and 10-year follow-up times within the NHANES dataset, which contains 47,261 samples and 151 features.

Addressing racial disparities in surgical care with machine learning
J Halamka et al, NPJ Digital Medicine, September 30, 2022 (Posted: Oct 03, 2022 6AM)

While inequalities will require numerous cultural, ethical, and sociological solutions, artificial intelligence-based algorithms may help address the problem by detecting bias in the data sets currently being used to make medical decisions. However, such AI-based solutions are only in early development.

Measurement of Exhaled Volatile Organic Compounds as a Biomarker for Personalised Medicine: Assessment of Short-Term Repeatability in Severe Asthma
A Azim et al, J Per Medicine, September 29, 2022 (Posted: Oct 02, 2022 9AM)

The intra-subject and between-subject variability of each VOC was calculated across the 70 samples and identified 30.35% of VOCs to be erratic: variable between subjects but also variable in the same subject. Exclusion of these erratic VOCs from machine learning approaches revealed no apparent loss of structure to the underlying data or loss of relationship with salient clinical characteristics. Moreover, cluster evaluation by the silhouette coefficient indicates more distinct clustering.

Can Smartphones Help Predict Suicide?
E Barry, NY Times, September 30, 2022 (Posted: Oct 01, 2022 7AM)

In the field of mental health, few new areas generate as much excitement as machine learning, which uses computer algorithms to better predict human behavior. There is, at the same time, exploding interest in biosensors that can track a person’s mood in real time, factoring in music choices, social media posts, facial expression and vocal expression. A unique research project is tracking hundreds of people at risk for suicide, using data from smartphones and wearable biosensors to identify periods of high danger — and intervene.

Predicting use of Intensive Care Units during the COVID-19 Pandemic
K Perez et al, Research Square, September 30, 2022 (Posted: Oct 01, 2022 7AM)

We propose ten approaches to represent the dynamics and predict the number of additional ICU beds required in the future. Machine learning models and classical time series analysis algorithms allow upper and lower bounds to be set for the number of units needed. Evaluating the predictions with 2020 and 2021 data in three representative geographic regions produces lower errors in the largest of the regions.

A machine learning approach identifies unresolving secondary pneumonia as a contributor to mortality in patients with severe pneumonia, including COVID-19
CA Gao et al, MEDRXIV, September 25, 2022 (Posted: Sep 27, 2022 8AM)

Integrated multimodal artificial intelligence framework for healthcare applications
LR Soenksen et al, NPJ Digital Medicine, September 20, 2022 (Posted: Sep 21, 2022 7AM)

AI systems leveraging multiple data sources and input modalities are poised to become a viable method to deliver more accurate results and deployable pipelines across a wide range of applications. In this work, we propose and evaluate a unified Holistic AI in Medicine (HAIM) framework to facilitate the generation and testing of AI systems that leverage multimodal inputs. Our approach uses generalizable data pre-processing and machine learning modeling stages that can be readily adapted for research and deployment in healthcare environments.

Machine Learning Algorithms for Prediction of Survival by Stress Echocardiography in Chronic Coronary Syndromes
L Cortiginai et al, J Per Med, September 16, 2022 (Posted: Sep 18, 2022 5AM)

Stress echocardiography (SE) is based on regional wall motion abnormalities and coronary flow velocity reserve (CFVR). Their independent prognostic capabilities could be better studied with a machine learning (ML) approach. The study aims to assess the SE outcome data by conducting an analysis with an ML approach. We included 6881 prospectively recruited and retrospectively analyzed patients with suspected (n = 4279) or known (n = 2602) coronary artery disease.

Steps to avoid overuse and misuse of machine learning in clinical research
V Volovici et al, Nature Medicine, September 12, 2022, (Posted: Sep 12, 2022 1PM)

Machine learning algorithms are a powerful tool in healthcare, but sometimes perform no better than traditional statistical techniques. Steps should be taken to ensure that algorithms are not overused or misused, in order to provide genuine benefit for patients. The lackluster performance of many machine learning (ML) systems in healthcare has been well documented. In healthcare, as in other areas, AI algorithms can even perpetuate human prejudices such as sexism and racism when trained on biased datasets.

Subpopulation-specific machine learning prognosis for underrepresented patients with double prioritized bias correction
S Afrose et al, Comm Medicine, September 1, 2022 (Posted: Sep 01, 2022 2PM)

Biases exist in the widely accepted one-machine-learning-model-fits-all-population approach. We invent a bias correction method that produces specialized machine learning prognostication models for underrepresented racial and age groups. This technique may reduce potentially life-threatening prediction mistakes for minority populations.


Disclaimer: Articles listed in Hot Topics of the Day are selected by Public Health Genomics Branch to provide current awareness of the scientific literature and news. Inclusion in the update does not necessarily represent the views of the Centers for Disease Control and Prevention nor does it imply endorsement of the article's methods or findings. CDC and DHHS assume no responsibility for the factual accuracy of the items presented. The selection, omission, or content of items does not imply any endorsement or other position taken by CDC or DHHS. Opinion, findings and conclusions expressed by the original authors of items included in the Clips, or persons quoted therein, are strictly their own and are in no way meant to represent the opinion or views of CDC or DHHS. References to publications, news sources, and non-CDC Websites are provided solely for informational purposes and do not imply endorsement by CDC or DHHS.