Express Healthcare

IIITH and NIMS Hyderabad release open-source histopathological datasets for brain cancer and kidney disease

India pathology dataset initiative digitises tissue biopsy slides to support AI-driven clinical research and education

0 217

IIIT Hyderabad (IIITH), in collaboration with Nizam’s Institute of Medical Sciences (NIMS), Hyderabad, has released publicly available datasets comprising digitised histopathological images of brain cancer and kidney disease (Lupus Nephritis). This initiative is part of the India Pathology Dataset (IPD) project, which involves academia, hospitals, industry, and the government. The project aims to digitise tissue biopsy slides to reduce the risk of damage to physical slides, improve clinical decision-making, accelerate turnaround times, and create research opportunities through artificial intelligence.

As part of the project supported by iHub-Data, IIITH installed a whole slide digital scanner at the NIMS campus. Prof. Vinod P.K, who is leading the dataset curation, explained, “Traditionally, tissue samples and biopsies are visualised under the microscope. But by digitising these slides, computers can be used to visualise these images, and they can be shared across locations for a collaborative diagnosis with other pathologists.”

One of the datasets released is the IPD-Brain dataset, published in Nature Scientific Data. This open-access dataset focuses on Indian demographics and comprises 547 high-resolution H&E slides from 367 patients, making it one of the largest in Asia. Dr Megha Uppin from the Department of Pathology at NIMS stated, “The effective management of all cancers relies on precise typing, sub-typing, and grading. This dataset is the first step in brain tumour research where machine learning models can be trained to explore regional and ethnic disease variations and enhance diagnostic precision.”

Dr Uppin noted that the diagnosis of brain tumours now largely relies on molecular genetics, and AI can bridge gaps in cost-effective and accurate diagnosis by predicting molecular abnormalities. She also mentioned that AI could address the shortage of specialised neuropathologists by enabling peripheral institutes to collaborate with experts through digital pathology. Efforts are underway to expand the dataset to include other cancers, such as breast, lung, colorectal, oral, and cervical cancers, with NIMS contributing to the lung cancer dataset.

In addition to cancer datasets, the project has compiled a dataset on lupus nephritis, a kidney disease caused by the immune system attacking the kidneys. According to Prof. Vinod, lupus nephritis disproportionately affects women in India, with a high incidence in Telangana. The dataset aims to support nephropathologists, who are in short supply in the country, by aiding in the classification of disease and directing appropriate treatment plans. Dr Uppin highlighted that AI can overcome interobserver variations in class subtyping, improving diagnostic consistency.

The project also explores using AI to predict molecular markers from H&E slide images, which traditional histopathology cannot achieve. Prof. Vinod explained, “The pathologist can’t see what underlying molecular changes are happening at the DNA level, which gets reflected in tissue morphology.” The team has worked on predicting molecular details, such as IDH mutations, which are critical for diagnosing and prognosing brain tumour patients.

The open-source nature of the dataset provides a valuable resource for researchers, educators, and healthcare professionals. “This is one of the first few instances of open-source medical data from India for ‘human good’,” said Prof. Vinod. A second whole slide scanner has been installed at the IIITH campus, which is accessible to dental colleges and corporate hospitals. The datasets are also valuable tools for pathology students to gain an in-depth understanding of histopathological images.

Prof. Vinod stated that more datasets, including one on breast cancer, are in progress. He emphasised that the project is tailored to India’s unique healthcare landscape, offering an alternative to datasets based on foreign populations. The IPD project aims to advance teaching, learning, and research in histopathology, providing a comprehensive database for healthcare innovation in India.

- Advertisement -

Leave A Reply

Your email address will not be published.