Timezone: »

 
Workshop
Machine Learning for Health (ML4H): Advancing Healthcare for All
Stephanie Hyland · Allen Schmaltz · Charles Onu · Ehi Nosakhare · Emily Alsentzer · Irene Y Chen · Matthew McDermott · Subhrajit Roy · Benjamin Akera · Dani Kiyasseh · Fabian Falck · Griffin Adams · Ioana Bica · Oliver J Bear Don't Walk IV · Suproteem Sarkar · Stephen Pfohl · Andrew Beam · Brett Beaulieu-Jones · Danielle Belgrave · Tristan Naumann

Fri Dec 11 06:00 AM -- 04:20 PM (PST) @ None
Event URL: https://ml4health.github.io/2020/ »

The application of machine learning to healthcare is often characterised by the development of cutting-edge technology aiming to improve patient outcomes. By developing sophisticated models on high-quality datasets we hope to better diagnose, forecast, and otherwise characterise the health of individuals. At the same time, when we build tools which aim to assist highly-specialised caregivers, we limit the benefit of machine learning to only those who can access such care. The fragility of healthcare access both globally and locally prompts us to ask, “How can machine learning be used to help enable healthcare for all?” - the theme of the 2020 ML4H workshop.

Participants at the workshop will be exposed to new questions in machine learning for healthcare, and be prompted to reflect on how their work sits within larger healthcare systems. Given the growing community of researchers in machine learning for health, the workshop will provide an opportunity to discuss common challenges, share expertise, and potentially spark new research directions. By drawing in experts from adjacent disciplines such as public health, fairness, epidemiology, and clinical practice, we aim to further strengthen the interdisciplinarity of machine learning for health.

See our workshop for more information: https://ml4health.github.io/

Fri 6:00 a.m. - 6:10 a.m.
Opening Remarks (Opening)
Fri 6:10 a.m. - 6:30 a.m.
Keynote 1: Noémie Elhadad (Keynote)
Noemie Elhadad
Fri 6:30 a.m. - 6:50 a.m.
Keynote 2: Mark Dredze (Keynote)
Mark Dredze
Fri 6:50 a.m. - 7:25 a.m.
Panel with keynotes 1-2 (Panel/QA)
Fri 7:25 a.m. - 7:40 a.m.
Break
Fri 7:40 a.m. - 8:00 a.m.
Sponsor remarks (Keynote)
Fri 8:00 a.m. - 8:10 a.m.
Spotlight A-1: "ML4H Auditing: From Paper to Practice" (Spotlights)
Luis Oala
Fri 8:10 a.m. - 8:20 a.m.
Spotlight A-2: "The unreasonable effectiveness of Batch-Norm statistics in addressing catastrophic forgetting across medical institutions" (Spotlights)
Sharut Gupta
Fri 8:20 a.m. - 8:30 a.m.
Spotlight A-3: "DeepHeartBeat: Latent trajectory learning of cardiac cycles using cardiac ultrasounds" (Spotlights)
Fabian Laumer
Fri 8:30 a.m. - 9:30 a.m.

https://gather.town/app/OVLlsG1Ar09XPYTN/ML4H2020

Trust Issues - Uncertainty Estimation Does not Enable Reliable OOD Prediction On Medical Tabular Data 1 Improved Clinical Abbreviation Expansion via Non-Sense-Based Approaches 2 Concept-based model explanations for Electronic Health Records 3 DeepHeartBeat: Sequence modelling for medical data 4 Temporal Pointwise Convolution Networks for Length of Stay Prediction in the ICU 5 Parkinsonian Chinese Speech Analysis towards Automatic Classification of Parkinson's Disease 6 Evaluation of Contrastive Predictive Coding for Histopathology Applications 7 Multiomics Data Analysis Predicts Risk of Preeclampsia 8 Confounding Feature Acquisition for Causal Effect Estimation 9 ParaMed: A Parallel Corpus for English-Chinese Translation in the Biomedical Domain 10 Natural Language Processing to Detect Cognitive Concerns in Electronic Health Records Using Deep Learning 11 Mobility network models of COVID-19 explain inequities and inform reopening 12 Incorporating Healthcare Motivated Constraints in Restless Multi-Armed Bandit Based Resource Allocation 13 Zero-Shot Clinical Acronym Expansion via Latent Meaning Cells 14 Decomposing Normal and Abnormal Features of Medical Images for Content-based Image Retrieval 15 Cost-Sensitive Machine Learning Classification for Mass Tuberculosis Verbal Screening 16 Uncertainty-Aware Counterfactual Explanations for Medical Diagnosis 17 Detecting small polyps using a Dynamic SSD-GAN 18 The unreasonable effectiveness of Batch-Norm statistics in addressing catastrophic forgetting across medical institutions 19 Learning transition times in event sequences: the Event-Based Hidden Markov Model of disease progression 20 Neural Temporal Point Processes For Modelling Electronic Health Records 21 Comparison of pharmacist evaluation of medication orders with predictions of a machine learning model 22 Medical symptom recognition from patient text: An active learning approach for long-tailed multilabel distributions 23 Accounting for Affect in Pain Level Recognition 24 Forecasting Emergency Department Capacity Constraints for COVID Isolation Beds 25 Spectral discontinuity design: Interrupted time series with spectral mixture kernels 26 A Bayesian Approach for Continual Learning in Clinical Time Series 27 A Neural SIR Model for Global Forecasting 28 MTB-HINE-BERT: a pre-trained genetic mutation representation model for predicting drug resistance of Mycobacterium tuberculosis 29 Stable predictions for health related anticausal prediction tasks affected by selection biases: the need to deconfound the test set features 30 Deep Learning Derived Histopathology Image Score for Increasing Phase 3 Clinical Trial Probability of Success 31 Self-Supervision Closes the Gap Between Weak and Strong Supervision in Histology 32 Model-Attentive Ensemble Learning for Sequence Modeling 33 Towards Automated Anamnesis Summarization: BERT-based Models for Symptom Extraction 34 Recommendations for Bayesian hierarchical model specifications for case-control studies in mental health 35 A decision-making tool to fine-tune abnormal levels in the complete blood count tests 36 Enhancing COVID-19 patient deterioration prediction using secondary task pre-trained embedding 37 Stratification of Systemic Lupus Erythematosus Patients with Gene Expression Data to Reveal Expression of Distinct Immune Pathways 38 Phenotyping Clusters of Patient Trajectories suffering from Complex Disease 39 Image analysis for Alzheimer's disease prediction: Embracing pathological hallmarks for model architecture design 40 Deep Cox Mixtures for Survival Regression 41

Fri 9:30 a.m. - 12:30 p.m.
Lunch (Break)
Fri 12:30 p.m. - 12:50 p.m.
Keynote 3: Judy Gichoya (Keynote)
Judy Gichoya
Fri 12:50 p.m. - 1:10 p.m.
Keynote 4: Ziad Obermeyer (Keynote)
Ziad Obermeyer
Fri 1:10 p.m. - 1:45 p.m.
Panel with keynotes 3-4 (Panel/QA)
Fri 1:45 p.m. - 1:55 p.m.
Spotlight B-1: "A Bayesian Hierarchical Network for Combining Heterogeneous Data Sources in Medical Diagnoses" (Spotlights)
Claire Donnat
Fri 1:55 p.m. - 2:05 p.m.
Spotlight B-2: "Assessing racial inequality in COVID-19 testing with Bayesian threshold tests" (Spotlights)
Emma Pierson
Fri 2:05 p.m. - 2:15 p.m.
Spotlight B-3: "EEG-GCNN: Augmenting Electroencephalogram-based Neurological Disease Diagnosis using a Domain-guided Graph Convolutional Neural Network" (Spotlights)
Neeraj Wagh
Fri 2:15 p.m. - 3:15 p.m.

https://gather.town/app/OVLlsG1Ar09XPYTN/ML4H2020

Towards Diagnosing Nonalcoholic Fatty Liver Disease with Abdominal MRI Data using Deep Learning 42 Adversarial Factor Models for Confound Disentangled Autism Biomarkers 43 Identifying Decision Points for Safe and Interpretable RL in Hypotension Treatment 44 Neural ODEs for Multi-State Survival Analysis 45 Assessing racial inequality in COVID-19 testing with Bayesian threshold tests 46 Learning to Predict and Support for Clinical Risk Stratification 47 sEMG Gesture Recognition with a Simple Model of Attention 48 Addressing the Real-world Class Imbalance Problem in Dermatology 49 Appropriate Evaluation of Diagnostic Utility of Machine Learning Algorithm Generated Images 50 Exploring Gender Disparities in Time to Diagnosis 51 Contrastive Representation Learning for Electroencephalogram Classification 52 Toward Understanding Clinical Context of Medication Change Events in Clinical Narratives 53 Pneumothorax and chest tube classification on chest x-rays for detection of missed pneumothorax 54 Interpretable Epilepsy Detection in Routine, Interictal EEG Data using Deep Learning 55 ML4H Auditing: From Paper to Practice 56 EEG-GCNN: Augmenting Electroencephalogram-based Neurological Disease Diagnosis using a Domain-guided Graph Convolutional Neural Network 57 GloFlow: Global Image Alignment for Creation of Whole Slide Images for Pathology from Video 58 A Bayesian Hierarchical Network for Combining Heterogeneous Data Sources in Medical Diagnoses 59 An Interpretable End-to-end Fine-tuning Approach for Long Clinical Text 60 A Study of Domain Generalization on Ultrasound-based Multi-Class Segmentation of Arteries, Veins, Ligaments, and Nerves Using Transfer Learning 61 CliniQG4QA: Generating Diverse Questions for Domain Adaptation of Clinical Question Answering 62 Download Source PDF Actions Copy Project Word Count Sync Dropbox Git GitHub Settings Compiler TeX Live version Main document Spell check Auto-complete Auto-close Brackets Code check Editor theme Overall theme Keybindings Font Size Font Family Line Height PDF Viewer Help Show Hotkeys Documentation Contact Us TL-Lite: Visualization and Temporal Learning for Clinical Cohorts 63 Estimating County-Level COVID-19 Exponential Growth Rates]{Estimating County-Level COVID-19 Exponential Growth Rates Using Generalized Random Forests 64 CheXphoto: 10,000+ Photos and Transformations of Chest X-rays for Benchmarking Deep Learning Robustness 65 Augmenting BERT Carefully with Underrepresented Linguistic Features 66 Towards Trainable Saliency Maps in Medical Imaging 67 transferGWAS: GWAS of Images Using Deep Transfer Learning 68 CheXphotogenic: Generalization of Deep Learning Models for Chest X-ray Interpretation to Photos of Chest X-rays 69 Generating SOAP Notes from Doctor-Patient Conversations 70 Personalized Healthcare and Causal Interventions 71 Attend and Decode: 4D fMRI Task State Decoding Using Attention Models 72 An Empirical Study of Representation Learning for Reinforcement Learning in Healthcare 73 Opening a Can of Words: Train-test overlaps in Clinical Natural Language Processing Datasets 74 A stability-driven protocol for drug response interpretable prediction (staDRIP) 75 Quantifying Common Support between Multiple Treatment Groups Using a Contrastive VAE 76 Learning Optimal Predictive Checklists 77 3D Photography Based Neural Network Craniosynostosis Triaging System 78 A Time-Aware Transformer Based Model for Suicide Ideation Detection on Social Media 79 Bayesian Recurrent State Space Model For rs-fMRI 80 COVID-19 in Differential Diagnosis of Online Symptom Assessments 81 Uncertainty-Aware Multi-Modal Ensembling for Severity Prediction of Alzheimer’s Dementia 82

Fri 3:15 p.m. - 3:30 p.m.
Break
Fri 3:30 p.m. - 3:50 p.m.
Keynote 5: Andrew Ng (Keynote)
Andrew Ng
Fri 3:50 p.m. - 4:10 p.m.
Panel with keynote 5 (Panel/QA)
Fri 4:10 p.m. - 4:20 p.m.
Closing remarks (Closing)

Author Information

Stephanie Hyland (Microsoft Research)
Allen Schmaltz (Harvard University)
Charles Onu (McGill University)
Ehi Nosakhare (Microsoft)
Emily Alsentzer (MIT)
Irene Y Chen (MIT)

Irene is a PhD student at MIT focusing on applications on health care and fairness. She did her undergrad at Harvard where I studied applied math and computational engineering. Before starting at MIT, she worked for two years at Dropbox as a data scientist and machine learning engineer.

Matthew McDermott (MIT)
Subhrajit Roy (Google)
Benjamin Akera (Mila - Quebec AI Institute)
Dani Kiyasseh (University of Oxford)
Fabian Falck (University of Oxford)
Griffin Adams (Columbia University)

I am an NLP researcher with a focus on text generation of clinical data. After completing a masters in Computational Data Science at Carnegie Mellon's Language Technologies Institute (LTI), I worked at Flatiron Health where I developed and deployed algorithms to extract clinical information from unstructured oncology data at scale. I introduced deep learning to the company and architected a generalized model that improves the status quo of information extraction from large-scale longitudinal clinical notes. I am now a computer science PhD student at Columbia University with Noemie Elhadad. My research focuses on controllable factual text generation of clinical narratives.

Ioana Bica (University of Oxford)
Oliver J Bear Don't Walk IV (Columbia University)
Suproteem Sarkar (Harvard)
Stephen Pfohl (Stanford University)
Andrew Beam (Harvard)
Brett Beaulieu-Jones (Harvard Medical School)
Danielle Belgrave (Microsoft Research)
Tristan Naumann (Microsoft Research)

More from the Same Authors