Every routine scan carries evidence of diseases no one thought to look for. I build models that read it.

I am a Ph.D. candidate in the Sabuncu Lab at Cornell University and Cornell Tech, advised by Prof. Mert R. Sabuncu. My work centers on multimodal learning for opportunistic screening: models that fuse imaging, physiological signals, and clinical text, stay reliable when a modality is missing, and decide which test is worth running next. I develop these methods with clinical collaborators at Weill Cornell Medicine and Columbia University Irving Medical Center.

Before Cornell, I completed my B.Sc. and M.Sc. in Biomedical Engineering at the Bangladesh University of Engineering and Technology (BUET), where I later served two years as a lecturer in the Department of Biomedical Engineering and am now on study leave.

Multimodal learning Vision–language models Medical image analysis Representation learning Reinforcement learning AI for radiology

News

  • ShapKO: Shapley-Adaptive Modality Knockout for Robust Multimodal Learning accepted at MICCAI 2026.
  • Passed my Ph.D. candidacy exam and received a non-thesis M.Sc. in Biomedical Engineering at Cornell.
  • The CT-LVEF study on detecting abnormal ejection fraction from non-contrast chest CT accepted at European Heart Journal – Digital Health.
  • Cardio Amyloid-AI accepted at European Heart Journal – Digital Health.
  • X-Cardia accepted at MIDL 2026 as an oral presentation (top 8%).
  • Attended the Duke & Chen Institute Joint Boot Camp for AI-Accelerated Medical Research at Duke University.
  • Talk on multimodal screening for transthyretin cardiac amyloidosis at NEBEC, NYU.
  • Poster at the Young Investigators' Award Competition, ACC New York Chapter.
  • Poster at the Cardiovascular Research Symposium, Weill Cornell Medicine.
  • Selected for the International Student Advisory Committee at Cornell Tech.
  • Started the Summer Immersion Program at Weill Cornell Medicine with Dr. Ashley N. Beecy.
  • Joined the Sabuncu Lab at Cornell Tech as a graduate research assistant.
  • Started my Ph.D. at Cornell University with a departmental fellowship.

Research

2026 –

Cost-aware diagnostic pathway optimization

A reinforcement learning agent that chooses the next diagnostic test under cost and accuracy constraints, producing patient-specific screening pathways rather than a fixed protocol.

Cornell Tech

2025 – 2026

Robust multimodal learning with missing modalities

A Shapley-adaptive modality knockout scheme that modulates training when one or more inputs are absent, improving robustness over standard fusion baselines.

Cornell Tech

2024 – 2026

Multimodal AI for cardiac amyloidosis screening

A deep learning system fusing CT imaging, ECG, and structured clinical features to detect ATTR cardiac amyloidosis in patients with severe aortic stenosis.

Cornell Tech & Weill Cornell Medicine

2023 – 2026

Opportunistic cardiac screening from routine chest CT

A framework that detects cardiac abnormalities in non-cardiac chest CT scans acquired for entirely unrelated indications.

Cornell Tech

2025

Cross-modal cardiac knowledge transfer

Phenotype-guided contrastive pre-training that aligns imaging with clinical text for vision–language models.

Cornell Tech

Publications

2026

  • 01
    ShapKO: Shapley-Adaptive Modality Knockout for Robust Multimodal Learning N.B. Nizam, F. Liu, S. Kwak, … M. Sabuncu MICCAIPaper
  • 02
    Cardio Amyloid-AI: Advanced Multimodal Screening for Transthyretin Cardiac Amyloidosis in Severe Aortic Stenosis Patients N.B. Nizam, A. Beecy, J. Groenendyk, F. Liu, … M. Sabuncu European Heart Journal – Digital HealthPaper
  • 03
    X-Cardia: Phenotype-Guided Cross-Modal Alignment for Opportunistic Cardiac Screening on Routine Chest CT N.B. Nizam, F. Liu, S. Kwak, … M. Sabuncu MIDLOral · top 8%Paper
  • 04
    HyperCT: Low-Rank Hypernet for Unified Chest CT Analysis F. Liu, S. Kwak, H. Phung, N.B. Nizam, … M. Sabuncu MIDLPaper
  • 05
    Beyond Machine Interpretation: Learning from Expert Over-Reads Improves ECG Diagnosis S. Kwak, F. Liu, N.B. Nizam, … M. Sabuncu MIDLPaper
  • 06
    RNED: Rotary Number Encoding and Decoding for Quantitative Medical VLM Analysis F. Liu, S. Kwak, N.B. Nizam, … M. Sabuncu CVPRPaper
  • 07
    An Artificial Intelligence Model to Detect Abnormal Ejection Fraction from Non-Contrast Chest Computed Tomography: The CT-LVEF Study J. Raikhelkar, Z. Bai, A. Beecy, R. Ilan, F. Liu, N.B. Nizam, V. Kishore, … N. Uriel European Heart Journal – Digital HealthPaper
  • 08
    BUET Multi-disease Heart Sound Dataset: A Comprehensive Auscultation Dataset for Developing Computer-Aided Diagnostic Systems S.N. Ali, A. Zahin, S.B. Shuvo, N.B. Nizam, … T. Hasan Computer Methods and Programs in Biomedicine UpdatePaper

2023

  • 09
    COVID-19 Severity Prediction from Chest X-ray Images using an Anatomy-Aware Deep Learning Model N.B. Nizam, S.M. Siddiquee, M. Shirin, M.I.H. Bhuiyan, T. Hasan Journal of Digital ImagingPaper

2022

  • 10
    Hilbert-Envelope Features for Cardiac Disease Classification from Noisy Phonocardiogram N.B. Nizam, S.I.S.K. Nuhash, T. Hasan Biomedical Signal Processing and ControlPaper
  • 11
    Anatomy X-Net: A Semi-Supervised Anatomy Aware Convolutional Neural Network for Thoracic Disease Classification U. Kamal, M. Zunaed, N.B. Nizam, T. Hasan IEEE Journal of Biomedical and Health InformaticsPaper
  • 12
    Bangla Speech-to-Braille Interaction Device for Visual and Hearing Impaired R.R. Rahim, A.I. Al Nahian, R.B. Kalam, A.S. Gupta, N.B. Nizam Springer FRAB ConferencePaper

2020

  • 13
    A Wavelet-CNN Feature Fusion Approach for Detecting COVID-19 from Chest Radiographs M.L. Rahman, N.B. Nizam, P. Datta, M.M. Hasan, T. Hasan, M.I.H. Bhuiyan IEEE ICECEPaper
  • 14
    Android Based Low Cost Sitting Posture Monitoring System N.B. Nizam, T. Jinan, W.B.N. Aurthy, M.R. Hossen, J. Ferdous IEEE ICECEPaper

Talks & presentations

  • 2026Shapley-Adaptive Modality Knockout for Robust Multimodal LearningMICCAI · Strasbourg, France
  • 2026X-Cardia: Phenotype-Guided Cross-Modal Alignment for Cardiac ScreeningMedical Imaging with Deep Learning (MIDL) · Taipei, Taiwan
  • 2025Multimodal Deep Learning for Opportunistic Cardiac Amyloidosis ScreeningYoung Investigators' Award Competition, ACC New York Chapter
  • 2025Opportunistic Screening of Cardiac AmyloidosisCardiovascular Research Symposium, Weill Cornell Medicine

Education & experience

  • 2023 – 2027Ph.D. in Biomedical EngineeringCornell University & Cornell Tech · Advisor: Prof. Mert R. Sabuncu
  • 2023 – 2026M.Sc. in Biomedical EngineeringCornell University & Cornell Tech
  • 2016 – 2023B.Sc. & M.Sc. in Biomedical EngineeringBUET, Dhaka · Advisor: Dr. Taufiq Hasan
  • 2024 – presentGraduate Research AssistantSabuncu Lab, Cornell Tech
  • 2021 – 2023Lecturer, Biomedical EngineeringBUET, Dhaka — on study leave
  • 2019 – 2021Undergraduate ResearchermHealth Lab, BUET
  • 2019Industrial TraineeSiemens, Dhaka
PythonC++MATLABPyTorchTensorFlow scikit-learnCUDAHPC / SLURMDICOM pipelinesAzureGit

Honors & awards

  • 20201st Runner-up, ACT COVID-19 National Call HackathonICT Ministry, Bangladesh
  • 20191st Runner-up, Johns Hopkins Healthcare Design Competition
  • 20191st Place, EEE Day Inter-University Project ShowBUET
  • 20192nd Best Presenter AwardBangladesh Cardiac Society Conference
  • 20192nd Runner-up, Design for Life CompetitionEdge Foundation, Bangladesh
  • 2016 – 2020Dean's List ScholarshipAll academic levels of undergraduate study, BUET
  • 2016 – 2020University Merit ScholarshipMultiple academic terms, BUET

Leadership & service

  • 2026 – presentStudent Mentor, Cornell TechMentoring undergraduates on research projects
  • 2024 – 2026International Student Advisory Committee, Cornell TechRepresenting international graduate students in policy and welfare discussions
  • 2018 – 2020Co-founder, GyanJamOnline learning platform reaching over 10,000 learners in Bangladesh