Network medicine predicts repurposable drugs for COVID-19.
problem Identifying effective drugs for SARS-CoV-2 infections quickly.
method Artificial intelligence, network diffusion, and network proximity algorithms.
result A multimodal approach combining predictions from multiple algorithms outperforms individual methods.
Automated image segmentation distinguishes overlapping human chromosomes.
problem Distinguishing overlapping human chromosomes for medical diagnostics.
method Customized convolutional neural network for image segmentation.
result IOU scores of 94.7% for overlapping regions, 88-94% for non-overlapping regions.
DeepCare models patient health history for predictive medicine.
problem Long-term temporal dependencies in patient illness and care processes.
method End-to-end deep dynamic neural network that reads medical records, stores history, infers current states, and predicts future outcomes.
result Improved disease progression modeling, intervention recommendation, and future risk prediction accuracy.
Deep neural networks are vulnerable to adversarial attacks in time series classification.
problem Vulnerability of deep learning models to adversarial time series examples.
method Proposed adversarial attack mechanisms to add noise to input time series.
result Current state-of-the-art deep learning time series classifiers are vulnerable to adversarial attacks.
New methods use network data to find genetic indicators for diseases.
problem Finding genetic indicators for diseases from large, complex data.
method Integrating network data to select features from whole-genome data.
result Methods can identify genetic indicators that work together.
Neural machine translation improves medical text quality.
problem Improving machine translation quality in medical text domains.
method Examined different training methods on a Polish-English medical text corpus.
result Neural machine translation outperformed traditional statistical methods.
Broad learning integrates diverse healthcare data for diagnostics and precision medicine.
problem Integrating various types of healthcare data for better diagnostics and personalized medicine.
method Fusing multi-view data including scalar, tensor, graph, and sequence data for knowledge discovery and machine learning tasks.
result Accurate user profiles and brain connectivity patterns can be created for improved diagnostics and personalized medicine.
Optimizes personalized medicine prediction by individual tuning.
problem Predicting individual drug responses using genomic information.
method Introduces a new ridge estimator and tuning parameter calibration scheme.
result Optimal in terms of oracle inequalities, fast, and highly effective.
The paper advocates for interpretable, accountable, reproducible machine learning in medicine.
problem Black box models in medicine lack transparency and regulatory approval.
method Intrinsically interpretable modeling approaches and collaborative learning paradigms.
result Interpretable machine learning models can support clinical decisions and gain regulatory approval.
Stein-Encoder isolates genetic signals in multi-modal biomedical data.
problem Integration of high-dimensional genomic data with clinical data obscures genetic predictive impact.
method White-box supervised framework using Stein's method and residualization.
result Stein-Encoder improves predictive accuracy and reveals specific biological mechanisms.
Convolutional neural network captures medical features for risk prediction in EHRs.
problem Extracting useful clinical representations from longitudinal EHR data.
method Multi-layer convolutional neural network (CNN) with learned medical feature embedding.
result Effective risk prediction on cohorts of congestive heart failure and diabetes patients.
Bandit algorithms optimize treatment decisions for precision medicine.
problem Optimizing treatment decisions for individual patients based on genetic or molecular profiling.
method Contextual bandit algorithms that consider patient characteristics.
result Bandit algorithms are useful for mobile health and digital phenotyping.
This paper reviews random forest methods for analyzing longitudinal data in precision medicine.
problem Analyzing longitudinal data for precision medicine.
method Extensions of random forest for longitudinal data analysis.
result Categorization of random forest methods for different data structures and repeated measurements.
New method optimizes individualized decision rules for precision medicine.
problem Heterogeneous patient responses to treatments.
method Proposes a decision-rule based optimized covariates dependent equivalent (CDE) for individualized decision making.
result Numerical experiments show improved performance in estimating optimal IDRs.
The validity of ML in medicine is compromised by inherent uncertainty in clinical data.
problem Inherent uncertainty in clinical data biases ML models in medicine.
method Analyzes the impact of uncertainty on ML models in medicine.
result Inherent uncertainty in clinical data undermines the clinical significance of ML outputs.
New method estimates composite outcomes for precision medicine.
problem Balancing multiple outcomes in precision medicine.
method Optimal treatment rule for composite outcomes.
result Estimates composite outcomes and optimal treatment rule.
The paper introduces a statistical framework for personalized medicine using microbiome data.
problem Analyzing high-dimensional microbiome data for personalized medicine.
method Phylogenetic LASSO method applied to microbiome data.
result A statistical framework for synthesizing personalized treatments from microbiome data.
AN2VEC disentangles feature and structural information in social networks.
problem Difficulty in separating feature and structural information in social networks.
method Graph Convolutional Networks (GCN) Variational Autoencoder for disentangled node embeddings.
result AN2VEC captures joint information of structure and features better than unshared information.
Bayesian network structures are usually built using only the data and starting from an empty network or from a naive Bayes structure. Very often, in some domains, like medicine, a prior structure knowledge is already known. This structure can be automatically or manually refined in search for better performance models.…
Learn embeddings from EHRs to predict ICD codes.
problem Predicting ICD codes from patient visits in EHRs.
method Deep neural network trained to predict ICD codes, capturing clinical information.
result Embeddings capture relevant clinical information and can be used in machine learning models.
Deep learning predicts patient trajectories in open Mimic-III dataset.
problem Predicting future medical conditions from patient history.
method Two parallel bi-directional Minimal Gated Recurrent Unit networks trained on Mimic-III dataset.
result Significant improvements in automated medical prognosis measured by Recall@k.
Generative models create personalized patient health simulations.
problem Creating accurate digital twins for personalized medicine.
method Neural network architecture for conditional generative models of clinical trajectories.
result Same architecture generates accurate twins across 13 indications.
Study compares WTT and DWT for FTIR data feature extraction of medicinal plants.
problem Improving machine learning efficiency with FTIR spectra of medicinal plants.
method Comparison of WTT and DWT for feature extraction, varying preprocessing steps.
result WTT and DWT yield similar results, improving clustering and classification accuracy.
SUBIC detects patient subgroups for personalized treatment of HTN in African-Americans.
problem Developing tailored treatment schemes for different patient subgroups.
method Supervised Biclustering using convex optimization.
result Identifies and prioritizes risk factors for HTN in African-Americans.
Study uses CNNs to estimate BMI from photos.
problem Estimating BMI from photos with limited data.
method Used deep learning (CNNs) to process silhouette images.
result High correlation between estimated and actual BMI.
3D dataset for intracranial aneurysms aids deep learning applications.
problem Lack of 3D medical datasets for deep learning.
method Developed an open-access 3D dataset, IntrA, for intracranial aneurysms.
result Demonstrated the challenges and performance of 3D medical datasets.
Deep learning estimates dose voxel kernels from CT data for personalized dosimetry.
problem Accurate estimation of absorbed dose in personalized radionuclide therapy.
method Combining deep learning with CT data to approximate dose voxel kernels from density kernels.
result Deep learning achieved an intersection-over-union score of 0.86 and mean squared error of 1.24×10−4 on real patient data.
Proposes deep feature selection for HTN risk factors in African-Americans.
problem Identifying significant risk factors for heart damage in African-Americans with hypertension.
method Uses stacked auto-encoders for feature learning and representation in deep architecture.
result Deep learning approach leads to better results in identifying LVMI risk factors.
Expands statistical background for knee osteoarthritis treatment models.
problem Developing optimal exercise and weight loss treatments for knee osteoarthritis.
method Precision medicine models and jackknife cross-validation method.
result Jackknife estimator provides consistent value function estimation.
AI-enabled precision medicine promises a transformational improvement in healthcare outcomes by enabling data-driven personalized diagnosis, prognosis, and treatment. However, the well-known "curse of dimensionality" and the clustered structure of biomedical data together interact to present a joint challenge in the hi…
Develops a Causal Transformer for estimating counterfactual outcomes from longitudinal data.
problem Estimating counterfactual outcomes over time from observational data is challenging due to complex, long-range dependencies.
method Combines three transformer subnetworks with separate inputs for time-varying covariates, previous treatments, and previous outcomes into a joint network with in-between cross-attentions. Uses a custom, end-to-end training procedure with a counterfactual domain confusion loss to address confounding bias.
result Achieves superior performance over current baselines in synthetic and real-world datasets.
P-Net predicts patient outcomes using sample networks.
problem Predicting patient outcomes based on sample data.
method P-Net constructs a network of patients, ranking and classifying them based on phenotype.
result P-Net performs similarly to classical methods in predicting patient outcomes.
Private RL algorithm with privacy guarantees for personalized medicine decisions.
problem Privacy-preserving reinforcement learning for personalized medicine decisions.
method Developed a private optimism-based RL algorithm using joint differential privacy (JDP).
result Achieved strong PAC and regret bounds with a privacy guarantee.
Proposes DSW for unbiased ITE estimation with dynamic confounders.
problem Estimating ITE from dynamic observational data with time-varying confounders.
method Deep Sequential Weighting (DSW) infers hidden confounders using current treatment assignments and historical information.
result DSW generates unbiased and accurate treatment effects.
Develops methods to learn optimal treatment regimes using causal tree methods.
problem Lack of methods for estimating treatment effects and handling complex patient data.
method Causal tree and causal forest methods for estimating heterogeneous treatment effects.
result Outperforms state-of-the-art baselines in cumulative regret and percentage of optimal decisions.
Estimates personalized models for individualized policies.
problem Localized effects are critical in precision medicine and advertising.
method Low-rank personalized regression with sample-specific models.
result Sample-specific models provide fine-grained interpretations without sacrificing predictive accuracy.
Survey on understanding neural networks for medical applications.
problem Black-box nature of deep neural networks hinders their use in critical applications.
method Comprehensive review of interpretability studies in neural networks.
result Interpretability research is crucial for the acceptance of neural networks in medical diagnosis.
Proposes a federated transfer learning method to improve precision medicine models for underrepresented populations.
problem Underrepresentation of minorities in precision medicine research leads to underperforming risk prediction models.
method Two-way federated transfer learning strategy integrating diverse populations and healthcare institutions.
result Improves risk prediction models for underrepresented populations, reducing performance gaps.
GraphMoE generates random graphs using neural networks and graphlets.
problem Learning generative models for random graphs.
method GraphMoE uses a neural network trained with graphlets and subgraph counts to match the distribution of random graphs.
result GraphMoE can generate graphs that mimic various real-world datasets and fool graph classifiers.
CNN learns diabetic syndromes from patient records.
problem Syndrome differentiation in TCM is complex and lacks standardization.
method Multi-instance multi-task learning with CNN.
result Outperforms other methods on diabetes dataset.
Neural network predicts falls in elderly people up to 10 minutes in advance.
problem Falls prevention in elderly people, especially in aging societies.
method Gated Recurrent Unit (GRU) based neural networks model using heart rate and mean blood pressure signals.
result Predicted syncope occurrence approximately 10 minutes before manual markers.
A new neural network reduces parameters for genomic data.
problem High-dimensional genomic data makes deep learning hard.
method Distributed representation and parameter prediction network.
result Significant reduction in parameters and error rate.
RIF prioritizes predictive biomarkers for precision medicine.
problem Lack of tools to select and prioritize predictive biomarkers.
method Random Interaction Forest (RIF) method.
result RIF outperformed conventional methods in various simulation scenarios and clinical trials.
PSICA identifies best treatments for patients with categorical therapies.
problem Identifying best treatments for patients with categorical therapies.
method Decision tree approach for subgroup identification in categorical treatment scenarios.
result Outputs a decision tree showing probabilities of best treatments for patient subgroups.
Parallel training speeds up neural network training, but process communication costs limit efficiency.
problem Efficiently training complex neural networks in real-time.
method Implemented Network Parallel Training using Cannon's Algorithm for matrix multiplication.
result Increasing the number of processes speeds up training until communication costs become prohibitive.
HAD-Net forecasts glucose levels with insights into insulin and carbs diffusion.
problem Inaccurate predictions in glucose level forecasting without context understanding.
method Hybrid model combining deep learning and physiological models, using recurrent attention network.
result Achieves competitive performance in glucose level forecasting with plausible diffusion insights.
Framework integrates mental disorder measurements for personalized treatment.
problem Optimizing treatment for mental disorders with latent mental states and heterogeneity.
method Measurement theory and multi-layer neural network for complex treatment effects.
result Learned treatment policies outperform alternatives on heterogeneous treatment effects.
AI enhances personalized drug development and decision-making in pharma.
problem Traditional drug development lacks personalized treatment plans.
method Application of AI in drug discovery, clinical trials, and post-marketing assessment.
result AI improves personalized medicine, optimizing health outcomes.