By Rajneesh Bhandari
Artificial Intelligence (AI), a technology increasingly adept at replicating human capabilities but with greater efficiency, speed, and cost-effectiveness, is poised to revolutionize healthcare. Potential applications of AI include diagnosis, image analysis, clinical trials, drug discovery, and patient interaction. As Microsoft’s CEO Satya Nadella said, “AI is perhaps the most transformational technology of our time, and healthcare is perhaps AI’s most pressing application.”
Studies have demonstrated that AI’s performance in image-based diagnoses meets or exceeds human experts in several medical specialties, including radiology, pathology, and dermatology. Around 90 percent of all healthcare data is derived from imaging technologies. AI offers faster, more accurate, and automated analysis. According to the American Cancer Society, many mammograms yield false positives, leading to 1 in 2 healthy women being incorrectly told they have cancer. AI can evaluate mammograms with 99 percent accuracy and 30 times faster, eliminating unnecessary biopsies and reducing patient anxiety.
AI-driven retinopathy screening can significantly reduce preventable vision loss related to diabetes. New AI-based software can complete MRI scans ten times faster, greatly enhancing patient comfort and reducing cost. Work is underway to use facial recognition software to identify rare diseases by detecting associated facial features. Researchers in South Korea have developed a method of diagnosing autism spectrum disorder (ASD) in children using retinal images screened by an AI algorithm. AI-assisted colonoscopy developed by Medtronics detects polyps and prevents colorectal cancer. Technologies like Zebra Medical Vision process millions of scans daily at minimal cost, suggesting a future where AI initially reviews medical images before being assessed by a radiologist.
Also read: Revolutionising public health delivery through AI use
Wearables continuously monitor health parameters such as ECG, oxygen saturation, sleep, gait, and heart rate. These devices are helpful only when integrated with AI to provide actionable insights to doctors and patients, facilitating early disease detection, enhancing patient adherence and motivation, and reducing long-term healthcare costs.
It takes an average of 12 years and up to a billion dollars to develop a new drug. Generative AI tools like Google’s AlphaFold accelerate drug development by generating novel molecular structures, swiftly screening compounds, predicting drug interactions, repurposing existing drugs, optimizing clinical trials, and enhancing drug formulations. According to Morgan Stanley, AI in drug development could create dozens of new medicines and a $50 billion market over the next decade.
In 2023, ChatGPT passed the US Medical Licensing Exam (USMLE), showcasing its potential in medical applications. Combined with human expertise, generative AI has led to virtual health assistants like Ada, which assesses symptoms, offers medical guidance in multiple languages, and has completed over 30 million consultations for 13 million patients.
Implementing AI in healthcare brings its own challenges. Data privacy, security, and regulatory concerns arise, especially for cloud-based AI. Regulating evolving AI systems that provide ongoing advice is complex, as approved AI can change over time and offer different insights.
AI in healthcare is about $11 billion market and is projected to reach $200 billion by 2030. India is uniquely positioned to lead AI healthcare with its strong IT sector, buzzing start-up ecosystem, and skilled professionals. India can drive significant advancements in healthcare solutions by investing in AI R&D, fostering public-private partnerships, and implementing supportive regulatory frameworks. Leveraging AI to address the country’s healthcare challenges, such as infectious diseases and rural healthcare delivery, can set a precedent for other developing nations.
(The author is Managing Director, NeuroEquilibrium. Views expressed are personal and do not reflect the official position or policy of FinancialExpress.com. Reproducing this content without permission is prohibited.)