Search

Background We previously developed and validated a predictive model to help clinicians identify hospitalized adults with coronavirus disease 2019 (COVID-19) who may be ready for discharge given their low risk of adverse events. Whether this algorithm can prompt more timely discharge for stable patients in practice is unknown. Objectives The aim of the study is to estimate the effect of displaying risk scores on length of stay (LOS). Methods We integrated model output into the electronic health record (EHR) at four hospitals in one health system by displaying a green/orange/red score indicating low/moderate/high-risk in a patient list column and a larger COVID-19 summary report visible for each patient. Display of the score was pseudo-randomized 1:1 into intervention and control arms using a patient identifier passed to the model execution code. Intervention effect was assessed by comparing LOS between intervention and control groups. Adverse safety outcomes of death, hospice, and re-presentation were tested separately and as a composite indicator. We tracked adoption and sustained use through daily counts of score displays. Results Enrolling 1,010 patients from May 15, 2020 to December 7, 2020, the trial found no detectable difference in LOS. The intervention had no impact on safety indicators of death, hospice or re-presentation after discharge. The scores were displayed consistently throughout the study period but the study lacks a causally linked process measure of provider actions based on the score. Secondary analysis revealed complex dynamics in LOS temporally, by primary symptom, and hospital location. Conclusion An AI-based COVID-19 risk score displayed passively to clinicians during routine care of hospitalized adults with COVID-19 was safe but had no detectable impact on LOS. Health technology challenges such as insufficient adoption, nonuniform use, and provider trust compounded with temporal factors of the COVID-19 pandemic may have contributed to the null result. Trial registration ClinicalTrials.gov identifier: NCT04570488.

The COVID-19 pandemic has challenged front-line clinical decision-making, leading to numerous published prognostic tools. However, few models have been prospectively validated and none report implementation in practice. Here, we use 3345 retrospective and 474 prospective hospitalizations to develop and validate a parsimonious model to identify patients with favorable outcomes within 96 h of a prediction, based on real-time lab values, vital signs, and oxygen support variables. In retrospective and prospective validation, the model achieves high average precision (88.6% 95% CI: [88.4–88.7] and 90.8% [90.8–90.8]) and discrimination (95.1% [95.1–95.2] and 86.8% [86.8–86.9]) respectively. We implemented and integrated the model into the EHR, achieving a positive predictive value of 93.3% with 41% sensitivity. Preliminary results suggest clinicians are adopting these scores into their clinical workflows.

Filtered through the analytical power of artificial intelligence, the wealth of available biomedical data promises to revolutionize cancer research, diagnosis and care. In this Viewpoint, six experts discuss some of the challenges, exciting developments and future questions arising at the interface of machine learning and oncology.

Early diagnosis of Alzheimer’s disease plays a pivotal role in patient care and clinical trials. In this study, we have developed a new approach based on 3D deep convolutional neural networks to accurately differentiate mild Alzheimer’s disease dementia from mild cognitive impairment and cognitively normal individuals using structural MRIs. For comparison, we have built a reference model based on the volumes and thickness of previously reported brain regions that are known to be implicated in disease progression. We validate both models on an internal held-out cohort from The Alzheimer’s Disease Neuroimaging Initiative (ADNI) and on an external independent cohort from The National Alzheimer’s Coordinating Center (NACC). The deep-learning model is more accurate and significantly faster than the volume/thickness model. The model can also be used to forecast progression: subjects with mild cognitive impairment misclassified as having mild Alzheimer’s disease dementia by the model were faster to progress to dementia over time. An analysis of the features learned by the proposed model shows that it relies on a wide range of regions associated with Alzheimer’s disease. These findings suggest that deep neural networks can automatically learn to identify imaging biomarkers that are predictive of Alzheimer’s disease, and leverage them to achieve accurate early detection of the disease.

Image-based analysis as a method for mutation detection can be advantageous in settings when tumor tissue is limited or unavailable for direct testing. In this study, we utilize two distinct and complementary machine-learning methods of analyzing whole-slide images for predicting mutated BRAF. In the first method, whole-slide images of melanomas from 256 patients were used to train a deep convolutional neural network to develop a fully automated model that first selects for tumor-rich areas (area under the curve = 0.96) and then predicts for mutated BRAF (area under the curve = 0.71). Saliency mapping was performed and revealed that pixels corresponding to nuclei were the most relevant to network learning. In the second method, whole-slide images were analyzed using a pathomics pipeline that first annotates nuclei and then quantifies nuclear features, showing that mutated BRAF nuclei were significantly larger and rounder than BRAF‒wild-type nuclei. Finally, we developed a model that combines clinical information, deep learning, and pathomics that improves the predictive performance for mutated BRAF to an area under the curve of 0.89. Not only does this provide additional insights on how BRAF mutations affect tumor structural characteristics, but machine learning‒based analysis of whole-slide images also has the potential to be integrated into higher-order models for understanding tumor biology.

Biopsies of inflammatory tissue contain a complex network of interacting cells, orchestrating the immune or autoimmune response. While standard histological examination can identify relationships, it is clear that a great amount of data on each slide is not quantitated or categorized in standard microscopic examinations. To deal with the huge amount of data present in biopsy tissue in an unbiased and comprehensive way, we have developed a deep learning algorithm to identify immune cells in biopsies of inflammatory lesions. We focused on T follicular helper (Tfh) cell subsets and B cells in dermatomyositis biopsy images. We achieved strong performance on detection and classification of cells, including the rare Tfh cell subsets present in the tissue. This algorithm could be used to perform distance mapping between cell types in tissue, and could be easily adapted to other disease states.

BACKGROUNDAlzheimer’s disease (AD) and Lewy body disease (LBD) are the two most common neurodegenerative dementias, and can occur in combination (AD+LBD). Due to overlapping biomarkers and symptoms, especially at early stages, clinical differentiation of these subtypes could be difficult. Yet it is unclear how the magnitude of diagnostic uncertainty varies across the dementia spectrum and demographic variables. We aimed to compare clinical diagnosis and post-mortem autopsy-confirmed pathological results to assess the clinical subtype diagnosis quality across these various factors.METHODSWe studied data from the National Alzheimer’s Coordinating Center. Selection criteria included an autopsy-based neuropathological assessment for AD and/or LBD, and initial visit with Clinical Dementia Rating CDR® Dementia Staging Instrument (CDR) of 0, 0.5, or 1.0 corresponding to normal, mild cognitive impairment, or mild dementia, respectively. Longitudinally, we analyzed first visits at each subsequent CDR stage, and focused on the clinical diagnosis for AD and/or LBD in these visits. This analysis included positive predictive values (PPV), specificity and sensitivity and false negative rates of the clinical diagnosis, as well as disparities in these metrics by sex, race/ethnicity, age and education. If autopsy-confirmed AD, LBD, or AD+LBD was missed, the alternative clinical diagnosis was analyzed.FINDINGSRecords of 1887 qualified participants from September 1, 2005 to August 20, 2019 were included in the study. Clinical diagnosis of AD+LBD had poor sensitivity from 2.7% (95% CI: 1.2%, 4.8%) at CDR 0.5 to 4.7% (2.1%, 7.8%) at CDR 3.0. Over 60% of participants with autopsy-confirmed AD+LBD were diagnosed clinically as AD. Clinical diagnosis of AD had low sensitivities at early dementia stage (56.8% (95% CI: 52.4%, 61.1%)) and low specificities at all stages. Among participants diagnosed as AD in the clinic, over 33% had concurrent LBD neuropathology at autopsy. Clinical diagnosis of LBD had poor PPVs from mild cognitive impairment (22% (95% CI: 12%, 33%)) to severe dementia (20% (95% CI: 9%, 33%)). Among participants diagnosed as LBD in the clinic, 32% to 54% revealed AD pathology during autopsy. When the three subtypes were missed by clinicians, “No cognitive impairment”, “Undecided” and “primary progressive aphasia or behavioral variant frontotemporal dementia” were the leading primary etiologic clinical diagnosis. With increasing dementia stages, the difference in clinical diagnosis accuracy significantly increased between White and Black/African-American participants, and diagnosis quality significantly improved for males but not for females.INTERPRETATIONOur findings demonstrate that clinical diagnosis of AD, LBD and AD+LBD are inaccurate and harbor significant disparities based on race and sex. These findings have important implications for clinical management, anticipatory guidance, trial enrollment, and applicability of potential disease modifying therapies for AD. They also promote research into better biomarker-based assessment of LBD pathology.

Electronic Health Records (EHR) are high-dimensional data with implicit connections among thousands of medical concepts. These connections, for instance, the co-occurrence of diseases and lab-disease correlations can be informative when only a subset of these variables is documented by the clinician. A feasible approach to improving the representation learning of EHR data is to associate relevant medical concepts and utilize these connections. Existing medical ontologies can be the reference for EHR structures, but they place numerous constraints on the data source. Recent progress on graph neural networks (GNN) enables end-to-end learning of topological structures for non-grid or non-sequential data. However, there are problems to be addressed on how to learn the medical graph adaptively and how to understand the effect of the medical graph on representation learning. In this paper, we propose a variationally regularized encoder-decoder graph network that achieves more robustness in graph structure learning by regularizing node representations. Our model outperforms the existing graph and non-graph based methods in various EHR predictive tasks based on both public data and real-world clinical data. Besides the improvements in empirical experiment performances, we provide an interpretation of the effect of variational regularization compared to standard graph neural network, using singular value analysis.

Summary The determination of endometrial carcinoma histological subtypes, molecular subtypes, and mutation status is critical for the diagnostic process, and directly affects patients’ prognosis and treatment. Sequencing, albeit slower and more expensive, can provide additional information on molecular subtypes and mutations that can be used to better select treatments. Here, we implement a customized multi-resolution deep convolutional neural network, Panoptes, that predicts not only the histological subtypes but also the molecular subtypes and 18 common gene mutations based on digitized H&E-stained pathological images. The model achieves high accuracy and generalizes well on independent datasets. Our results suggest that Panoptes, with further refinement, has the potential for clinical application to help pathologists determine molecular subtypes and mutations of endometrial carcinoma without sequencing., Graphical abstract, Highlights CNN models predict subtypes and mutations in endometrial cancer based on H&E images Multi-resolution CNN models perform better on H&E images than single-resolution ones Feature extraction suggests CNN models use human interpretable tumor features Tumor grade distinguishes CNV-H molecular subtype from endometrioid histology samples, Hong et al. develop and implement a customized multi-resolution deep convolutional neural network that predict molecular subtypes and 18 common gene mutations in endometrial cancer based on digitized H&E-stained pathological images. The models learn human interpretable and generalizable features, indicating potential clinical application without sequencing analysis.

Twitter data has been shown broadly applicable for public health surveillance. Previous public heath studies based on Twitter data have largely relied on keyword-matching or topic models for clustering relevant tweets. However, both methods suffer from the short-length of texts and unpredictable noise that naturally occurs in user-generated contexts. In response, we introduce a deep learning approach that uses hashtags as a form of supervision and learns tweet embeddings for extracting informative textual features. In this case study, we address the specific task of estimating state-level obesity from dietary-related textual features. Our approach yields an estimation that strongly correlates the textual features to government data and outperforms the keyword-matching baseline. The results also demonstrate the potential of discovering risk factors using the textual features. This method is general-purpose and can be applied to a wide range of Twitter-based public health studies.

During the coronavirus disease 2019 (COVID-19) pandemic, rapid and accurate triage of patients at the emergency department is critical to inform decision-making. We propose a data-driven approach for automatic prediction of deterioration risk using a deep neural network that learns from chest X-ray images and a gradient boosting model that learns from routine clinical variables. Our AI prognosis system, trained using data from 3661 patients, achieves an area under the receiver operating characteristic curve (AUC) of 0.786 (95% CI: 0.745–0.830) when predicting deterioration within 96 hours. The deep neural network extracts informative areas of chest X-ray images to assist clinicians in interpreting the predictions and performs comparably to two radiologists in a reader study. In order to verify performance in a real clinical setting, we silently deployed a preliminary version of the deep neural network at New York University Langone Health during the first wave of the pandemic, which produced accurate predictions in real-time. In summary, our findings demonstrate the potential of the proposed system for assisting front-line physicians in the triage of COVID-19 patients.

Visual inspection of histopathology slides is one of the main methods used by pathologists to assess the stage, type and subtype of lung tumors. Adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC) are the most prevalent subtypes of lung cancer, and their distinction requires visual inspection by an experienced pathologist. In this study, we trained a deep convolutional neural network (inception v3) on whole-slide images obtained from The Cancer Genome Atlas to accurately and automatically classify them into LUAD, LUSC or normal lung tissue. The performance of our method is comparable to that of pathologists, with an average area under the curve (AUC) of 0.97. Our model was validated on independent datasets of frozen tissues, formalin-fixed paraffin-embedded tissues and biopsies. Furthermore, we trained the network to predict the ten most commonly mutated genes in LUAD. We found that six of them—STK11, EGFR, FAT1, SETBP1, KRAS and TP53—can be predicted from pathology images, with AUCs from 0.733 to 0.856 as measured on a held-out population. These findings suggest that deep-learning models can assist pathologists in the detection of cancer subtype or gene mutations. Our approach can be applied to any cancer type, and the code is available at https://github.com/ncoudray/DeepPATH .

Recently a lot of work has been done on applying deep learning to medical problems. In this paper, we use Brain MRI images to predict parameters like memory scores, body mass index (BMI) and age of a person with simple 3D neural network architectures. We model the problem as both single and multi task. We compare the performance of the two approaches. We also use visualizations to find potential biomarkers for these parameters.

Visual analysis of solid tissue mounted on glass slides is currently the primary method used by pathologists for determining the stage, type and subtypes of cancer. Although whole slide images are usually large (10s to 100s thousands pixels wide), an exhaustive though time-consuming assessment is necessary to reduce the risk of misdiagnosis. In an effort to address the many diagnostic challenges faced by trained experts, recent research has been focused on developing automatic prediction systems for this multi-class classification problem. Typically, complex convolutional neural network (CNN) architectures, such as Google’s Inception, are used to tackle this problem. Here, we introduce a greatly simplified CNN architecture, PathCNN, which allows for more efficient use of computational resources and better classification performance. Using this improved architecture, we trained simultaneously on whole-slide images from multiple tumor sites and corresponding non-neoplastic tissue. Dimensionality reduction analysis of the weights of the last layer of the network capture groups of images that faithfully represent the different types of cancer, highlighting at the same time differences in staining and capturing outliers, artifacts and misclassification errors. Our code is available online at:https://github.com/sedab/PathCNN.

Image-based analysis as a rapid method for mutation detection can be advantageous in research or clinical settings when tumor tissue is limited or unavailable for direct testing. Here, we applied a deep convolutional neural network (CNN) to whole slide images of melanomas from 256 patients and developed a fully automated model that first selects for tumor-rich areas (Area Under the Curve AUC=0.96) then predicts for the presence of mutated BRAF in our test set (AUC=0.72) Model performance was cross-validated on melanoma images from The Cancer Genome Atlas (AUC=0.75). We confirm that the mutated BRAF genotype is linked to phenotypic alterations at the level of the nucleus through saliency mapping and pathomics analysis, which reveal that cells with mutated BRAF exhibit larger and rounder nuclei. Not only do these findings provide additional insights on how BRAF mutations affects tumor structural characteristics, deep learning-based analysis of histopathology images have the potential to be integrated into higher order models for understanding tumor biology, developing biomarkers, and predicting clinical outcomes.

[This corrects the article DOI: 10.1371/journal.pone.0215571.].

Importance To date, the association of psychiatric diagnoses with mortality in patients infected with coronavirus disease 2019 (COVID-19) has not been evaluated. Objective To assess whether a diagnosis of a schizophrenia spectrum disorder, mood disorder, or anxiety disorder is associated with mortality in patients with COVID-19. Design, Setting, and Participants This retrospective cohort study assessed 7348 consecutive adult patients for 45 days following laboratory-confirmed COVID-19 between March 3 and May 31, 2020, in a large academic medical system in New York. The final date of follow-up was July 15, 2020. Patients without available medical records before testing were excluded. Exposures Patients were categorized based on the followingInternational Statistical Classification of Diseases, Tenth Revision, Clinical Modificationdiagnoses before their testing date: (1) schizophrenia spectrum disorders, (2) mood disorders, and (3) anxiety disorders. Patients with these diagnoses were compared with a reference group without psychiatric disorders. Main Outcomes and Measures Mortality, defined as death or discharge to hospice within 45 days following a positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) test result. Results Of the 26 540 patients tested, 7348 tested positive for SARS-CoV-2 (mean [SD] age, 54 [18.6] years; 3891 [53.0%] women). Of eligible patients with positive test results, 75 patients (1.0%) had a history of a schizophrenia spectrum illness, 564 (7.7%) had a history of a mood disorder, and 360 (4.9%) had a history of an anxiety disorder. After adjusting for demographic and medical risk factors, a premorbid diagnosis of a schizophrenia spectrum disorder was significantly associated with mortality (odds ratio [OR], 2.67; 95% CI, 1.48-4.80). Diagnoses of mood disorders (OR, 1.14; 95% CI, 0.87-1.49) and anxiety disorders (OR, 0.96; 95% CI, 0.65-1.41) were not associated with mortality after adjustment. In comparison with other risk factors, a diagnosis of schizophrenia ranked behind only age in strength of an association with mortality. Conclusions and Relevance In this cohort study of adults with SARS-CoV-2–positive test results in a large New York medical system, adults with a schizophrenia spectrum disorder diagnosis were associated with an increased risk for mortality, but those with mood and anxiety disorders were not associated with a risk of mortality. These results suggest that schizophrenia spectrum disorders may be a risk factor for mortality in patients with COVID-19.

Machine learning has recently gained considerable attention because of promising results for a wide range of radiology applications. Here we review recent work using machine learning in brain tumor imaging, specifically segmentation and MRI radiomics of gliomas.We discuss available resources, state-of-the-art segmentation methods, and machine learning radiomics for glioma. We highlight the challenges of these techniques as well as the future potential in clinical diagnostics, prognostics, and decision making.

Suchi Saria of Johns Hopkins University shares how big data and machine learning can help improve the practice of healthcare, and how computing students can contribute.

Visual analysis of histopathology slides of lung cell tissues is one of the main methods used by pathologists to assess the stage, types and sub-types of lung cancers. Adenocarcinoma and squamous cell carcinoma are two most prevalent sub-types of lung cancer, but their distinction can be challenging and time-consuming even for the expert eye. In this study, we trained a deep learning convolutional neural network (CNN) model (inception v3) on histopathology images obtained from The Cancer Genome Atlas (TCGA) to accurately classify whole-slide pathology images into adenocarcinoma, squamous cell carcinoma or normal lung tissue. Our method slightly outperforms a human pathologist, achieving better sensitivity and specificity, with ∼0.97 average Area Under the Curve (AUC) on a held-out population of whole-slide scans. Furthermore, we trained the neural network to predict the ten most commonly mutated genes in lung adenocarcinoma. We found that six of these genes – STK11, EGFR, FAT1, SETBP1, KRAS and TP53 – can be predicted from pathology images with an accuracy ranging from 0.733 to 0.856, as measured by the AUC on the held-out population. These findings suggest that deep learning models can offer both specialists and patients a fast, accurate and inexpensive detection of cancer types or gene mutations, and thus have a significant impact on cancer treatment.

BackgroundBecause of the strong link between childhood obesity and adulthood obesity comorbidities, and the difficulty in decreasing body mass index (BMI) later in life, effective strategies are needed to address this condition in early childhood. The ability to predict obesity before age five could be a useful tool, allowing prevention strategies to focus on high risk children. The few existing prediction models for obesity in childhood have primarily employed data from longitudinal cohort studies, relying on difficult to collect data that are not readily available to all practitioners. Instead, we utilized real-world unaugmented electronic health record (EHR) data from the first two years of life to predict obesity status at age five, an approach not yet taken in pediatric obesity research.Methods and findingsWe trained a variety of machine learning algorithms to perform both binary classification and regression. Following previous studies demonstrating different obesity determinants for boys and girls, we similarly developed separate models for both groups. In each of the separate models for boys and girls we found that weight for length z-score, BMI between 19 and 24 months, and the last BMI measure recorded before age two were the most important features for prediction. The best performing models were able to predict obesity with an Area Under the Receiver Operator Characteristic Curve (AUC) of 81.7% for girls and 76.1% for boys.ConclusionsWe were able to predict obesity at age five using EHR data with an AUC comparable to cohort-based studies, reducing the need for investment in additional data collection. Our results suggest that machine learning approaches for predicting future childhood obesity using EHR data could improve the ability of clinicians and researchers to drive future policy, intervention design, and the decision-making process in a clinical setting.

We present a new approach to population health, in which data-driven predictive models are learned for outcomes such as type 2 diabetes. Our approach enables risk assessment from readily available electronic claims data on large populations, without additional screening cost. Proposed model uncovers early and late-stage risk factors. Using administrative claims, pharmacy records, healthcare utilization, and laboratory results of 4.1 million individuals between 2005 and 2009, an initial set of 42,000 variables were derived that together describe the full health status and history of every individual. Machine learning was then used to methodically enhance predictive variable set and fit models predicting onset of type 2 diabetes in 2009-2011, 2010-2012, and 2011-2013. We compared the enhanced model with a parsimonious model consisting of known diabetes risk factors in a real-world environment, where missing values are common and prevalent. Furthermore, we analyzed novel and known risk factors emerging from the model at different age groups at different stages before the onset. Parsimonious model using 21 classic diabetes risk factors resulted in area under ROC curve (AUC) of 0.75 for diabetes prediction within a 2-year window following the baseline. The enhanced model increased the AUC to 0.80, with about 900 variables selected as predictive (p 0.0001 for differences between AUCs). Similar improvements were observed for models predicting diabetes onset 1-3 years and 2-4 years after baseline. The enhanced model improved positive predictive value by at least 50% and identified novel surrogate risk factors for type 2 diabetes, such as chronic liver disease (odds ratio [OR] 3.71), high alanine aminotransferase (OR 2.26), esophageal reflux (OR 1.85), and history of acute bronchitis (OR 1.45). Liver risk factors emerge later in the process of diabetes development compared with obesity-related factors such as hypertension and high hemoglobin A1c. In conclusion, population-level risk prediction for type 2 diabetes using readily available administrative data is feasible and has better prediction performance than classical diabetes risk prediction algorithms on very large populations with missing data. The new model enables intervention allocation at national scale quickly and accurately and recovers potentially novel risk factors at different stages before the disease onset.

To characterize a tumor and suggest appropriate treatments, pathologists rely on the visual assessment of histopathology slides as well as molecular profiling of the tumors. This is especially true in lung adenocarcinoma cases where targeted therapies exist for tumors with EGFR mutations. In this work, we studied the use of deep learning approaches to classify lung cancer histopathology images and predict the mutational status of frequently mutated genes. To achieve these tasks, more than 1600 scanned histopathology slides, obtained from TCGA, were separated into independent training, validation and test sets. They were then tiled into 512x512 pixel windows, and the training set was used to feed the inception v3 convolutional neural network developed by Google. The resulting model was evaluated on the test set, and the generated per-tile probabilities were aggregated to generate a per-slide classification score. We based our study on whole-slide images of adenocarcinoma, squamous cells carcinoma and normal lung tissues and the neural network was trained on several tasks: first, the goal was to identify normal from tumor regions, and then, once the tumor regions were identified, it was trained to identify adenocarcinoma versus squamous cell carcinoma. Finally, the neural network was trained predict the mutational status of the most frequently mutated genes in lung adenocarcinoma. After verifying that the trained network properly identifies tumor slides compared to normal tissue (AUC=0.99) and distinguishes adenocarcinomas from squamous cell carcinomas (AUC=0.95), the network was trained to simultaneously identify different mutations. It achieved AUC scores of 0.82 and 0.86 for STK11 and EGFR respectively, showing that mutations in these two genes may confer specific macroscopic features, visible in histopathology images, and that a deep convolutional neural network can be trained to recognize those features. Our results demonstrate that, given the availability of sufficient data, convolutional neural networks learn to perform complex diagnoses and thus assist pathologists in their work. Citation Format: Nicolas Coudray, Andre L. Moreira, Theodore Sakellaropoulos, David Fenyö, Narges Razavian, Aristotelis Tsirigos. Determining EGFR and STK11 mutational status in lung adenocarcinoma histopathology images using deep learning [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5309.