The Indian Institute of Technology (IIT) has a group of researchers working on this project (IIT)
Eye images captured by low-cost near-infrared (NIR) cameras have been discovered to aid in low design costs, the convenience of use, and practical solutions for cataract detection in Jodhpur. In short, the AI-based approach could be beneficial in cataract surgery or other procedures, making them more inexpensive for individuals.
The suggested multitask deep learning algorithm is known as MTCD
It is low-cost and produces extremely high levels of precision. A deep learning-based cataract diagnosis approach involving iris segmentation and multitask network classification is presented in this study. The suggested segmentation technique “detects non-ideal ocular borders fast and effectively,” the researchers claim. The suggested multitask deep learning algorithm is known as MTCD.
The proposed strategy can be applied in rural areas
Cataracts are typically identified via fundus pictures, which are more expensive to acquire and require expertise to operate the fundus cameras. It may be made more accessible and affordable by combining an AI-based solution with low-cost imaging sensors. It is also cost-effective because it uses low-cost NIR cameras instead of expensive ophthalmoscopes. According to the research, the proposed strategy can be applied in rural areas where doctors are few.
The team
Dr. Mayank Vatsa and Dr. Richa Singh of IIT Jodhpur’s Image Analysis and Biometrics (IAB) Lab conceptualised the study. Mahapara Khurshid, Yasmeena Akhter, Rohit Keshari, Pavani Tripathi, and Aditya Lakra were among the UG and Ph.D. students in the lab who assisted them.
Currently, a big number of patients with cataracts have to visit secondary and tertiary care centers,” Dr. Richa Singh, professor, department of computer science and engineering, IIT Jodhpur, said of the study. The existence of such a solution can enable clinicians at basic health care clinics to aid such patients.”
“We are extending this research to include both cataract and diabetic retinopathy in the solution,” said Dr. Mayank Vatsa, Professor, Department of Computer Science and Engineering, IIT Jodhpur. “We have collaborated with multiple hospitals in the country for domain expertise, data collection, and validation of the solution.”
