Early disease detection made easier with AI

This article is authored by Shibu Vijayan, medical director, Global Health, Qure.ai.

Even as India’s health care system has progressed immensely over the years, the majority of patients succumb to death because of the late detection of disease. It’s a challenge that the country is still facing. According to the National Academy of Medical Sciences (NAMS) report, most people in India only visit a hospital or see a doctor when the disease has progressed to an advanced stage or when it is too late. I believe this is a very crucial time to fully embrace the power of Artificial Intelligence (AI) in health care for India–especially in the early detection and diagnosis of diseases.

Artificial intelligence.(Thinkstock)

India is on the right path, with a substantial increase in AI expenditure, reaching $665 million in 2018 and projected to reach $11.78 billion by 2025, according to data from NITI Aayog. Successful collaborations between AI companies and the health care department further emphasise this trend.

NITI Aayog is exploring AI in primary care to detect diabetes complications early. They are also validating AI as a screening tool in eye care, comparing its accuracy with that of retina specialists. By integrating AI with portable screening devices like 3Nethra, eye screenings and early detection can be expanded, benefiting remote areas in India.

In a 2019 experiment by Google developers Artificial Intelligence and its future potential in lung cancer screening showed an AI algorithm that could detect lung cancer from CT scans with high accuracy. The algorithm outperformed radiologists in identifying malignant lung nodules, potentially enabling early detection and intervention.

Similarly, a study published in Nature Medicine in 2020 demonstrated the effectiveness of AI in detecting lung cancer from low-dose CT scans. The AI model achieved a sensitivity of 94% in detecting cancerous nodules, highlighting its potential for early diagnosis.

Additionally, research conducted by the Indian Council of Medical Research (ICMR) showcases the potential of AI in diagnosing tuberculosis. By examining chest X-rays and CT scans, AI algorithms can detect tuberculosis with high accuracy, enabling early intervention and better treatment outcomes.

On another note, a study published in the National Library of Medicine named Artificial Intelligence Models in the Diagnosis of Adult-Onset Dementia Disorders shows that the association of AI approaches with Magnetic Resonance Imaging (MRI) is particularly useful for improving the diagnostic accuracy of different dementia types. The application of AI techniques in conjunction with MRI has yielded a notable improvement in diagnostic accuracy, ranging from 73.3% to 99%. These findings indicate that integrating AI with conventional MRI techniques can enable a more precise and early diagnosis of dementia disorders.

Researchers have also developed AI models that can predict the risk of heart failure in individuals. These models analyse various data sources, including electronic health records, medical imaging, laboratory results, and patient demographics, to identify individuals at higher risk of developing heart failure. AI has been employed to enhance the analysis of cardiac imaging modalities like cardiac MRI and CT scans.

Despite these promising findings, India has yet to fully tap into the vast potential of AI in early disease detection. The NAMS Task Force Report on AI in health care highlights several challenges within India's health care system. These include a growing ageing population, limited access to health care, outdated record-keeping systems, inconsistent health care quality, and a shortage of skilled health care workers. These challenges must be addressed to fully leverage AI in early disease detection.

Besides strengthening the robust digital infrastructure, NITI Aayog’s National Strategy for Artificial Intelligence is working toward harnessing the power of AI to the fullest in the country. At the same time, I believe it would be in the best interest of the national population if the government encourages the sharing of health care data among institutions while ensuring patient privacy and data security. Establishing standards for data collection, storage, and interoperability will facilitate the integration of diverse datasets, enabling more comprehensive and representative AI models.

Also, this is the right time to invest in skill development programs to create a workforce with experience in both AI and health care. The initiatives could include training programmes, workshops, and collaborations between academic institutions and industry partners to bridge the skill gap. Building a skilled workforce can take India a long way in its AI journey.

Public-private partnerships play a vital role in driving innovation, knowledge sharing, and the development of AI-driven solutions. By leveraging shared expertise and resources, we can accelerate progress and overcome barriers that hinder early disease detection. The G20 Health care Summit is also, in my view, a great opportunity to discuss the use cases of AI in early disease detection.

I believe spreading awareness is the key here. The government should spearhead awareness campaigns targeting health care professionals, policymakers, and the general public. Educating stakeholders about the benefits and limitations of AI in disease detection will foster acceptance and facilitate its integration into the health care ecosystems.

The potential of AI in early disease detection is quite promising. By addressing challenges related to data accessibility, health care infrastructure, ethical frameworks, workforce training, and validation, India can embrace the health care revolution brought forth by AI. With the right approach, early disease detection powered by AI can lead to improved patient outcomes, reduced health care burdens, and a healthier future for India.

This article is authored by Shibu Vijayan, medical director, Global Health, Qure.ai.