A multidisciplinary team of doctors at the Post Graduate Institute of Medical Education and Research (PGIMER), led by Dr Pankaj Gupta, has developed an artificial intelligence (AI)-based application that can detect gallbladder cancer (GBC) using ultrasound images. The innovation is expected to aid early diagnosis, especially in regions lacking specialist radiologists.According to researchers, the tool can be of great use in peripheral hospitals, most of which lack specialised experts, including radiologists, to examine ultrasound images, do further line of investigation, CT scan, liver functioning tests, etc. (HT File)The study, titled ‘Multiple instance learning approach for automated gallbladder cancer detection using ultrasound imaging: multi-center validation of a deep learning model’, was published in The Lancet Regional Health - Southeast Asia in April 2026.How it worksWhile gallstones are easily identified through ultrasound, subtle early signs of gallbladder cancer can sometimes be missed even by experts. The AI model, developed by PGIMER, will be able to tell whether or not the patient is suffering from gallbladder cancer by analysing multiple ultrasound images. The app will also have a probability score to indicate the confidence level in the diagnosis. It will highlight the regions in the ultrasound images that influenced its assessment, enabling doctors to understand and verify the findings. It was tested on patients from four hospitals across North India.According to researchers, the tool can be of great use in peripheral hospitals, most of which lack specialised experts, including radiologists, to examine ultrasound images, do further line of investigation, CT scan, liver functioning tests, etc.India accounts for 10% of global burdenIndia has a high incidence rate of gallbladder cancer, contributing to about 10% of the global burden. The disease has a particularly high incidence in north, north-east, central and eastern India compared to the southern and western regions, according to a paper by Dr Usha Dutta, head of the gastroenterology department at PGIMER, titled ‘Epidemiology of Gallbladder Cancer in India’ published in Chinese Clinical Oncology.Most common digestive cancer among North Indian womenThe ICMR-National Institute of Cancer Prevention and Research describes gallbladder cancer as the most common digestive cancer among North Indian women. Recent PGIMER data revealed the incidence among women in North India is 21 per one lakh population, while among men it is less than 10 per one lakh. Around 80% of gallbladder cancer patients also suffer from gallstones.To be shared with hospitals free of costThe user-friendly computer application was developed by computer scientist Kartik Bose. Dr Pankaj Gupta said the team has been working on the project since 2018 and is now sharing the application free of cost with hospitals of national importance that have requested access. The team plans to further validate the model through prospective clinical trials and integrate it into routine ultrasound workflows. “The ultimate goal is to make AI-assisted gallbladder cancer screening accessible to healthcare providers across India, particularly in regions with high disease burden,” Dr Gupta said.