New specialized adjunct facilities in India’s premier healthcare institutions are propelling the advancement of precision medicine in the country through the utilization of artificial intelligence (AI).
Recently unveiled at the Apollo Cancer Centre (ACC) in Bengaluru, the potentially pioneering Precision Oncology Centre, empowered by AI, delivers individualized and comprehensive specialized care to each patient.
This cutting-edge center incorporates AI-driven automation to pinpoint suitable candidates for targeted therapy and immunotherapy, as well as to promptly alert healthcare teams of any patient deterioration. Moreover, it integrates conversational AI to educate patients and their families on diagnoses, treatments, and connections to support groups. Furthermore, the facility employs AI to oversee adherence to standard care protocols, facilitate patient management based on genomic, clinical, and pathological data, and provide recommendations for diagnostic tests and enrollment in value-based care initiatives and other patient benefit programs.
ACC asserts that the integration of AI, coupled with the analysis of vast health data, represents the “future of oncology,” ensuring precise diagnoses, real-time insights, cancer risk evaluations, treatment strategies, and continuous care.
Meanwhile, leveraging its AI expertise, the Indian Institute of Science (IISc) has inaugurated a collaborative AI laboratory in Precision Medicine in partnership with Siemens Healthineers, a reputable name in medical imaging. This laboratory aims to develop open-source AI tools that automatically segment pathological observations in brain scans. These tools, soon to be integrated into routine clinical processes, are designed to aid in the accurate diagnosis of neurological disorders and assess their clinical implications at a population level.
Drs. Vijay Agarwal and Vishwanath S, esteemed Medical Oncology senior consultants at ACC, shared with Healthcare IT News intricate details regarding their AI applications in precision oncology. Vaanathi Sundaresan, assistant professor at IISc’s Department of Computational and Data Sciences and head of the Siemens Healthineers-Computational Data Sciences Collaborative Laboratory for AI in Precision Medicine, discussed the measures taken to safeguard sensitive patient data amidst escalating cybersecurity risks.
Q: Can you elaborate on specific AI applications within your new facility?
Dr. Agarwal, ACC: An illustrative case involves a female patient who presented with a breast lump seeking consultation. Through AI, she swiftly received a breast cancer diagnosis within 24 hours. Subsequent to the diagnosis, an automated alert was triggered for all stakeholders – the attending physician, chief breast surgeon, multidisciplinary team (MDT) coordinator, and the patient – prompting an MDT consultation. Following the MDT meeting, treatment recommendations were relayed to the center, initiating the treatment process. The patient’s care pathway was predefined using AI, ensuring all stakeholders were informed. Seamless coordination across all specialties was achieved through automated alerts for chemotherapy initiation, admission procedures, medication orders, prescription administration, drug dispensing, consent management, discharge processes, and billing, thereby enhancing efficiency and reducing costs. Any modifications to the treatment plan were automatically communicated to all involved parties, streamlining care delivery. Subsequently, the patient received chemotherapy, targeted therapy, and was eventually referred to the MDT.
Dr. Vishwanath S, ACC: AI facilitates the seamless administration of chemotherapy from registration and bed allocation to discharge. It also supports personalized therapy based on NGS (next-generation sequencing) mutation status. Additionally, AI can drive digital pathology and imaging analyses – for instance, utilizing bioinformatics and AI to identify a patient with advanced sarcoma who has undergone extensive prior treatments and exhibits a targetable mutation as per the NGS report.
A/Prof Sundaresan, IISc: Other critical AI applications essential for clinical implementation include modeling disease progression at a population level, adapting AI models to variations in data characteristics across different sites (domain adaptation), addressing data scarcity scenarios, limited manual labeling availability, and outlier detection. A significant future direction involves exploring the correlation between brain health and other bodily organs.
Q: What is the inaugural project planned for the collaborative lab? How crucial is the demand for precise imaging/diagnosis of neurological diseases, and how can AI contribute to this?
A/Prof Sundaresan: Our primary project focuses on identifying vascular biomarkers in neuroimaging data to facilitate early detection of neurodegeneration.
Neurodegenerative ailments (such as Alzheimer’s disease and various forms of dementia) and cerebrovascular conditions like stroke are linked to cognitive impairment, gait abnormalities, brain atrophy, and, in severe cases, fatality (with stroke mortality rates reaching up to 47%). These conditions are frequently observed in individuals with vascular risk factors and depression. AI algorithms applied to MRI scans can detect imaging biomarkers critical for personalized treatment. However, the differential diagnosis and long-term prognosis of such neurological disorders necessitate precise imaging biomarkers and thorough assessment of their clinical implications, where AI methodologies can offer valuable support.
Q: Considering the lab’s extensive reliance on sensitive data, how do you plan to safeguard this data and the algorithms/models you will employ?
A/Prof Sundaresan: Initial experiments in the lab will utilize publicly available data for testing purposes. Any clinical data obtained for research (from IISc or collaborators) will be acquired post ethical board approvals, strictly anonymized, and privacy protected. The methods (excluding training data) will be shared as open-source for the broader research community’s benefit.
