Federated Learning for Diabetes Prediction: Investigating Client-Level Statistical Heterogeneity
ID:217 View Protection:ATTENDEE Updated Time:2026-11-19 02:26:41 Hits:40 In-person

Start Time:2026-11-20 15:00(Asia/Kolkata)

Duration:15min

Session:S4 Artificial Intelligence and Intelligent Computing » S4-1Artificial Intelligence and Intelligent Computing

Presentation File

Tips: Only the registered participant can access the file. Please sign in first.

Abstract
Machine learning can flag diabetes risk early, but there are privacy issues related to  centralized training, which depends  on  pooling  patient  records  by  hospitals.  Federated learning  trains  the  global  model  without  transferring  records to  a  central  server,  although  how  class  imbalance,  difference in  client  datasets,  and  the  number  of  clients  influence  this technique are not known. This paper compares centralized and federated  logistic  regression  on  96,146  records  of  a  publicly available  diabetes  dataset.  The  problem  of  class  imbalance  is handled by applying SMOTENC locally for each client, training with FedProx, and aggregation based on the sample size. One threshold of 0.8752, which maximizes the F1-score for out-of- fold  predictions,  is  used  for  all  models.  The  clients  are  split into IID or non-IID (Dirichlet, alpha of 0.30) partitions for 3, 5,  and  10  clients  across  ten  seeds.  All  federated  IID  models perform equally with the centralized baseline with F1-score close to 72.45 percent in all experiments. In case of non-IID splits, ROC-AUC is 0.955-0.959 against 0.960, while the mean F1-score is  68.70  percent  with  3  clients  and  recall  is  35  to  39  percent in two partitions. Calibration analysis attributes this to clients having insufficient diabetic records to sample from, which leads to under-prediction by the global model; across 30 non-IID runs, the effective positive share correlates with recall at 0.95 but not with ROC-AUC. FedProx coefficients of 0.05 to 0.1 decreased the variability relative to FedAvg.
Keywords
Federated Learning,Diabetes Prediction,logistic regression,class imbalance,SMOTENC,FedProx,Client heterogeneity
Speaker
Sanjana R
Student Rajalakshmi Engineering College

Submission Author
Aswana Lal Rajalakshmi Engineering College
Prathyush R Rajalakshmi Engineering College
Sanjana R Rajalakshmi Engineering College
Srevarsha N Rajalakshmi Engneering College
Naveed Ahmed Basha Rajalakshmi Engineering College
Submit Comment
Verify Code Change Another
All Comments
Important Date
  • Conference Date

    Nov 20

    2026

    to

    Nov 22

    2026

  • Oct 31 2026

    Draft paper submission deadline

  • Nov 07 2026

    Early Bird Registration

  • Nov 22 2026

    Registration deadline

Sponsored By
United Societies of Science
Organized By
Pondicherry University
Supported By
IEEE Section
IEEE Madras Section