Clinical AI models fail to transfer between sites due to site-specific practices.
problem Clinical AI models perform poorly at new sites.
method Identify and isolate site-specific clinical practices affecting data distribution.
result A potential solution to improve model transferability.
The study compares machine learning models for depression detection and highlights the importance of feature selection.
problem The choice of features in machine learning models for depression detection is crucial.
method Comparison of seven machine learning models on depression detection tasks.
result Optimal feature selection is essential for accurate and clinically acceptable classification solutions.
Deep neural networks improve ECG diagnosis accuracy.
problem Limited accuracy of existing ECG analysis models.
method Trained a deep neural network on a large dataset of 12-lead ECG exams.
result DNN outperforms human doctors in recognizing 6 types of ECG abnormalities.
Over half a million individuals are diagnosed with head and neck cancer each year worldwide. Radiotherapy is an important curative treatment for this disease, but it requires manual time consuming delineation of radio-sensitive organs at risk (OARs). This planning process can delay treatment, while also introducing int…
Semi-supervised learning method augments minority class examples for robust anomaly detection in clinical signals.
problem Class imbalance in minority class instances impairs robustness of clinical analytics solutions.
method Intelligent augmentation of minority class examples to balance class distribution and construct a smooth decision boundary.
result The proposed method outperforms state-of-the-art algorithms in anomaly detection for clinical signals.
Deep RL optimizes lab test scheduling for better patient outcomes and cost savings.
problem Redundant lab tests lead to cost and patient discomfort.
method Deep reinforcement learning for optimal scheduling.
result Deep RL policy outperforms heuristic scheduling in both accuracy and cost.
SEEDA optimizes dose allocation in clinical trials to balance efficacy and safety.
problem Complex relationships between efficacy and toxicity in new drug trials.
method Adaptive clinical trial methodology that maximizes cumulative efficacy while ensuring safety constraints.
result SEEDA outperforms existing methods in finding optimal doses with higher success rates and fewer patients.
Study identifies key aspects of explainable ML for clinical trust.
problem Lack of concrete definitions for usable explanations in clinical settings.
method Surveyed clinicians from two specialties to understand their needs for explainability.
result Characterized specific aspects of explainability that improve trust in ML models.
Objectives: Electronic health records (EHRs) are only a first step in capturing and utilizing health-related data - the challenge is turning that data into useful information. Furthermore, EHRs are increasingly likely to include data relating to patient outcomes, functionality such as clinical decision support, and gen…
Study improves healthcare time series imputation by considering structured missingness.
problem Structured missingness in clinical data impacts time series imputation models.
method Analysis of different masking strategies on imputation methods using PhysioNet Challenge 2012 dataset.
result Masking choices significantly affect imputation accuracy and clinical prediction.
Unsupervised learning improves clinical predictions from medical time series.
problem Improving clinical decision making through unlabeled medical data.
method Evaluation of unsupervised representation learning on medical time series using sequence-to-sequence models.
result A forecasting Seq2Seq model with an attention mechanism achieves the best performance.
System exposes study population descriptions in clinical guidelines.
problem Challenges in understanding applicability of clinical guidelines.
method Developed an ontology-enabled prototype system using SIO.
result Allows medical practitioners to better understand study populations.
Study examines biases in clinical word embeddings, revealing performance gaps across groups.
problem Biases in clinical word embeddings leading to performance differences across groups.
method Pretrained BERT models on MIMIC-III, fill-in-the-blank method, fairness evaluation on clinical tasks.
result Classifiers trained from BERT representations exhibit statistically significant differences in performance across groups.
Analysis of flow cytometry data is an essential tool for clinical diagnosis of hematological and immunological conditions. Current clinical workflows rely on a manual process called gating to classify cells into their canonical types. This dependence on human annotation limits the rate, reproducibility, and complexity …
Compressed sensing improves MRI scans with data-driven learning.
problem Challenges in applying compressed sensing from research to clinical practice.
method Data-driven learning to address challenges of hand-crafted priors, tuning parameters, and long reconstruction times.
result Compressed sensing can have greater clinical impact with data-driven learning.
Novel 3D U-Net method for fast, reproducible white matter tract segmentation.
problem Challenges in fast and consistent white matter tract segmentation from diffusion tensor MRI.
method Convolutional neural network (3D U-Net) trained on a large DTI dataset.
result Reproducibility and accuracy of tract-specific diffusion measures.
We developed an automated deep learning system to detect hip fractures from frontal pelvic x-rays, an important and common radiological task. Our system was trained on a decade of clinical x-rays (~53,000 studies) and can be applied to clinical data, automatically excluding inappropriate and technically unsatisfactory …
Clinical models trained on EHRs degrade in performance over time due to data drift.
problem Model performance degradation over time in clinical settings.
method Accessed year of care for each record in MIMIC, aggregated features into clinical concepts, and tested mitigation strategies.
result State-of-the-art models show significant performance drops when tested on future data compared to historical data.
Deep learning analyzes healthcare provider actions and patient outcomes.
problem Understanding provider behavior in non-randomized healthcare settings.
method Deep causal behavioral policy learning (DC-BPL) using transformer architecture.
result Optimal provider policies identified for specific patient types.
AACE learns treatment policies from EHRs using annotations to improve accuracy.
problem Learning treatment policies from multimodal EHRs with bias and inefficiency.
method Annotation-assisted coarsened effects (AACE) method.
result AACE outperforms existing methods in predicting treatment benefit from multimodal EHRs.
Stability in clinical prediction models is crucial for transferability between studies, yet has received little attention. The problem is paramount in high dimensional data which invites sparse models with feature selection capability. We introduce an effective method to stabilize sparse Cox model of time-to-events usi…
SARD improves deep learning clinical prediction performance.
problem Deep learning models struggle to match linear models in healthcare predictions.
method Reverse Distillation to initialize deep models, combined with contextual and temporal embeddings.
result SARD outperforms state-of-the-art methods on clinical prediction outcomes.
Develops ML-DQA for healthcare data quality assurance.
problem Inconsistent use of real-world data in machine learning projects.
method Develops ML-DQA framework based on RWD best practices.
result Five generalizable practices emerge from ML-DQA implementation.
Machine learning guides clinicians in predictive modeling using big data.
problem Insufficient understanding of machine learning among clinicians hinders its adoption.
method Provides a series of guides on machine learning principles, resampling, model evaluation, and coding.
result Clinicians need methodological rigor and clarity to use machine learning effectively.
CNNs can be trained with clinically available segmentations for OARs in radiotherapy.
problem Lack of dedicated training volumes for CNNs in radiotherapy.
method Used clinically available segmentations from PACS, applied multi-label segmentation, empirically assessed training set size.
result Clinically available segmentations can be used to train an accurate OAR segmentation model.
Paper proposes machine learning model for early Alzheimer's diagnosis.
problem Early and accurate diagnosis of Alzheimer's Disease.
method Machine learning models, demographic, biomarker, and cognitive test data.
result 90% accuracy and 87% accuracy in predicting Alzheimer's development.
New accuracy measure Ha improves AI system assessment in clinical practice.
problem Inadequate metrics for assessing AI system performance in clinical settings.
method Introducing H-accuracy (Ha) as a more informative measure.
result H-accuracy is a generalization of balanced accuracy and related to Net Benefit.
DWTS uses observational data to improve clinical trial efficiency.
problem Lack of definitive conclusions from randomized clinical trials due to insufficient patient cohorts and confounding biases.
method DWTS combines observational data with randomized clinical trials using Doubly Debiased LASSO (DDL) to identify reliable covariates.
result DWTS reduces cumulative regret in clinical trials compared to standard methods.
AdaCare learns health status from biomarkers across multiple time scales.
problem Lack of explicit extraction of historical biomarker variation and adaptability to diverse patient conditions.
method Scale-adaptive feature extraction and recalibration for interpretability.
result AdaCare achieves state-of-the-art prediction accuracy and provides interpretable results.
New method stabilizes deep learning models for clinical risk prediction.
problem Stability issues in deep learning models for clinical risk prediction.
method Bootstrapping-based regularisation framework embedded in deep neural networks.
result Improved prediction stability across multiple datasets.
Review of machine learning methods for detecting depression from resting EEG.
problem Detecting depression from resting-state EEG recordings.
method Machine learning combined with nonlinear analysis on EEG data.
result Improved reliability of depression detection models.
Machine learning models trained on hospital data degrade over time due to changing practices.
problem Machine learning models trained on hospital data degrade over time due to changing practices.
method We augmented MIMIC with the year of care and showed that a model trained using standard feature representations will significantly degrade in quality over time.
result Clinically-oriented aggregates of raw features significantly mitigate future deterioration.
QC methods improve reliability of machine learning-based image segmentation.
problem Inaccuracies in machine learning algorithms limit their clinical applicability.
method Analysis and validation of QC approaches for automatic segmentation.
result Aggregation of uncertainty and Dice prediction methods improved segmentation reliability.
Neural model predicts survival outcomes and reveals feature relationships.
problem Predicting time-to-event outcomes and understanding feature relationships in clinical data.
method Survival and topic modeling combined in a neural network framework.
result Neural survival-supervised topic models achieve competitive accuracy with interpretability.
Teaches reproducible research to medical students and postgrads.
problem Lack of reproducibility in medical research practices.
method Designed and delivered a lecture series on reproducible research.
result Encountered practical obstacles in reproducing a published analysis.
The availability of a large amount of electronic health records (EHR) provides huge opportunities to improve health care service by mining these data. One important application is clinical endpoint prediction, which aims to predict whether a disease, a symptom or an abnormal lab test will happen in the future according…
Study estimates treatment effect on survival outcomes using targeted maximum likelihood estimation.
problem Estimating treatment effect on time-to-event outcomes in clinical settings.
method Divided into three phases: estimation, feature selection, and targeted maximum likelihood estimation.
result Method performs well in high sample size or event rate conditions.
New diagnostics detect variability in individual risk estimates from machine learning models in healthcare.
problem Variability in individual risk estimates from machine learning models in healthcare, leading to unreliable treatment decisions.
method Proposed evaluation framework using empirical prediction interval width and empirical decision flip rate diagnostics.
result Randomness in optimization and initialization can lead to substantial individual-level variability in risk estimates, affecting clinical decisions.
The study evaluates AI model performance measures for medical use.
problem Selecting appropriate performance measures for AI models in medical practice.
method Assessed 32 performance measures across five domains for binary outcomes.
result 17 measures are both proper and reflect decision-analytic performance.
In this work we explored building automatic speech recognition models for transcribing doctor patient conversation. We collected a large scale dataset of clinical conversations (14,000 hr), designed the task to represent the real word scenario, and explored several alignment approaches to iteratively improve data qua…
New dataset from clinicians improves sepsis prediction models.
problem Circularity in previous sepsis prediction models.
method Developed an independent dataset from clinical judgments, avoiding circularity.
result Achieved state-of-the-art AUROC scores.
Personalized models using group attributes reduce performance, study finds.
problem Reducing performance of models using group attributes like race or gender.
method Formal conditions and collective preference guarantees to ensure fair use.
result Models personalized with group attributes reduce performance at a group level.
pmsims R package uses Gaussian process for flexible sample size estimation in clinical models.
problem Determining adequate sample size for clinical prediction models.
method Simulation-based Gaussian process search for flexible sample size estimation.
result Gaussian process-based method produces more stable sample size estimates, especially in challenging settings.
We present algorithms for the detection of a class of heart arrhythmias with the goal of eventual adoption by practicing cardiologists. In clinical practice, detection is based on a small number of meaningful features extracted from the heartbeat cycle. However, techniques proposed in the literature use high dimensiona…
Novel MCA-based rule mining speeds up psychiatric disorder screening.
problem High dimensionality and heterogeneity in clinical data.
method Developed a novel categorical rule mining method based on MCA.
result 100x faster than state-of-the-art methods for large datasets.
Deep learning predicts ICU mortality with enhanced interpretability.
problem Improving mortality prediction accuracy and clinician trust in AI.
method Trained a deep learning model on MIMIC-III to interpret nursing notes.
result Model reaches ROC of 0.8629, outperforming SAPS-II.
Paper develops machine learning algorithms to learn optimal integer weights for clinical risk scores.
problem Deriving optimal integer weights for clinical risk scores without computational burden.
method Flexible greedy optimization strategy to directly optimize a value function.
result Constructed an integer-weighted comorbidity score for measuring post-discharge mortality risk.
Develops CSI for predicting disease progression from sparse data.
problem Lack of frequent clinical metrics for disease progression.
method Machine learning framework (CSI) for analyzing disease progression from sparse observations.
result CSI effectively predicts disease progression from sparse data.