Home / Current Issue / Paper 1717648
Intelligent Rare Medical Event Detection
Subject area: Science,Engineering and Technology · Area of research: Medical AI
DOI: https://doi.org/10.64388/IREV9I11-1717648
Abstract
Identifying rare diseases is one of the toughest hurdles in modern medicine be-cause information is scarce and symptoms are often confusingly diverse, which fre-quently leads to long and stressful diag-nostic delays for patients. To solve this, we created CliniFlow AI, an intelligent platform designed to help doctors spot these rare conditions much earlier using a unique "brain" called Anomaly-Aware Adaptive Multimodal Fusion (A²MF), al-lowing it to recognize over 230 rare dis-eases even when data is extremely lim-ited. The system connects the dots by ana-lyzing clinical text with BioBERT, spot-ting abnormalities in medical images with a customized ResNet-50, and tracking a patient's health history over time using LSTM networks. By flagging unusual pat-terns through Anomaly Detection and learning from a small number of examples via Few-Shot Learning, CliniFlow AI pro-vides doctors with clear risk levels and diagnostic insights, making it a powerful and easy-to-scale tool for real-world hos-pitals. Beyond simple automation, this framework acts as a second pair of eyes that stays sharp during high-stakes medi-cal screenings where every minute counts. It effectively bridges the gap between massive amounts of raw hospital data and the specialized, actionable knowledge cli-nicians need to save lives.
Keywords
Rare Disease Detection, Mul-timodal Learning, Clinical Decision Sup-port System, Medical Image Analysis, Bi-oBERT, Few-Shot Learning, Anomaly De-tection, Explainable AI.
How to cite this paper
@article{1717648,
author = {Soham Mhatre, Dr. Pratibha Adkar},
title = {Intelligent Rare Medical Event Detection},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {11},
pages = {1479-1485},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1717648.pdf},
abstract = {Identifying rare diseases is one of the toughest hurdles in modern medicine be-cause information is scarce and symptoms are often confusingly diverse, which fre-quently leads to long and stressful diag-nostic delays for patients. To solve this, we created CliniFlow AI, an intelligent platform designed to help doctors spot these rare conditions much earlier using a unique "brain" called Anomaly-Aware Adaptive Multimodal Fusion (A²MF), al-lowing it to recognize over 230 rare dis-eases even when data is extremely lim-ited. The system connects the dots by ana-lyzing clinical text with BioBERT, spot-ting abnormalities in medical images with a customized ResNet-50, and tracking a patient's health history over time using LSTM networks. By flagging unusual pat-terns through Anomaly Detection and learning from a small number of examples via Few-Shot Learning, CliniFlow AI pro-vides doctors with clear risk levels and diagnostic insights, making it a powerful and easy-to-scale tool for real-world hos-pitals. Beyond simple automation, this framework acts as a second pair of eyes that stays sharp during high-stakes medi-cal screenings where every minute counts. It effectively bridges the gap between massive amounts of raw hospital data and the specialized, actionable knowledge cli-nicians need to save lives.},
keywords = {Rare Disease Detection, Mul-timodal Learning, Clinical Decision Sup-port System, Medical Image Analysis, Bi-oBERT, Few-Shot Learning, Anomaly De-tection, Explainable AI.},
month = {May},
doi = {https://doi.org/10.64388/IREV9I11-1717648}
}