Mass General Brigham seeks a Postdoctoral Research Fellow to develop and translate Medical AI for human neurophysiology and clinical care, with emphasis on EEG and MEG, multimodal signals, and affective computing, to reshape critical care.
Requirements
- Possess or be on track to complete a PhD or MD with background in computer science, physics, computational science, machine learning, artificial intelligence, deep learning, computer vision, large language model, image processing, biomedical engineering, bioinformatics, visual science, biochemical engineering or a related quantitative science field.
- Research experience and/or publications in computer vision, large language models (LLMs), vision science, and biomedical signal processing—especially EEG waveform analysis.
- Experience with facial expression analysis, affective computing, and sentiment analysis.
- Strong interest and background in biomedicine, chemical engineering, neuroscience and clinical research, with hands-on neurophysiology analytics (EEG/EMG) experience: end-to-end processing pipelines, artifact handling, time–frequency features, source modeling, and/or deep sequence models.
- Experience with multimodal learning (e.g., EEG + video/voice/wearables); self-supervised or foundation models; interpretability (causal/counterfactual analyses); and on-device or real-time inference.
- Excellent scientific writing and presentation skills.
- Software product development experience in Python.
Responsibilities
- Invent and prototype machine learning models for medical time series and multimodal data (e.g., EEG/EHR/video/wearables): self-/semi-supervised learning, representation learning, sequence models (TCN/Transformer), multimodal fusion, uncertainty & calibration, domain adaptation, and active learning.
- Build robust preprocessing and feature pipelines (artifact removal/ICA, filtering, time–frequency transforms, connectivity/graph features) and benchmarking baselines.
- Design rigorous evaluation plans (nested CV; AUROC/AUPRC; sensitivity/specificity at fixed thresholds; calibration/decision-curve analysis; subgroup/fairness slices; robustness, drift, and ablation studies).
- Own data curation/QC, labeling protocols, and dataset versioning; create task definitions/labels that reflect clinical intent and minimize information leakage.
- Maintain clean, well-tested code; containers and experiment tracking; model/data cards; automated training/validation scripts.
- Turn findings into clear figures and narratives; present at lab meetings and conferences; lead/assist manuscripts to top venues.
- Partner with clinicians/biostatisticians on problem framing and endpoint selection; mentor students/RAs on analysis best practices.
Other
- Work independently under the general guidance of the Principal Investigator.
- Be self-motivated, meticulous, highly organized, and detail-oriented.
- Demonstrate initiative and a rapid ability to learn new methods and tools.
- Exhibit strong analytical thinking and problem-solving skills.
- Possess advanced technical aptitude and data science proficiency.
