Artificial intelligence can assist, not replace us in treating cardiovascular diseases: Cardiologists

The menace of cardiovascular diseases, especially heart attack continues to rise by the day. Actor Shreyash Talpade recently suffered a heart attack. This acted as a stark reminder yet again of how any one of us could be at risk of heart disease. Medical experts have been urging all to stay vigilant and take corrective measures well in time to ensure complete protection. In such times, the existence of AI or artificial intelligence is proving to be a boon.

Medical experts speak of how a simple digital watch helped patients identify irregular rhythm in their heartbeats and seek immediate help. This is a simple example of how artificial intelligence is on a path to transform global healthcare as well.

Many doctors today speak of the great potential of artificial intelligence revolutionising medical research, disease detection, diagnosis and treatment. The same applies to CVD or cardiovascular disease care.

With the rampant rise in the risk of heart disease, it is crucial to understand how AI can help in early detection and risk mitigation. To gain a simple understanding of a somewhat complex subject matter, Mid-day Online reached out to highly experienced medical experts and cardiologists who share relevant insights.

How AI helps in the early detection of CVD
"Artificial Intelligence plays a pivotal role in transforming cardiovascular disease (CVD) care. It excels in predicting and preventing CVD by analysing extensive patient data and incorporating factors like medical history, lifestyle, genetics, and biomarkers. This aids in identifying individualised risks early and implementing targeted preventive measures," elucidates Dr Pragya Khare, associate consultant (non-invasive preventive cardiology), Fortis Escorts Heart Institute.

Continuing, she says that AI revolutionises CVD detection via medical imaging analysis and predictive analytics. AI-powered software scrutinises MRIs, CT scans, and echocardiograms, accurately spotting heart structure changes and arterial plaque. Predictive models leverage patient data to foresee CVD risks, aiding in early intervention planning. "Recent breakthroughs include DeepMind's AI analysing eye scans to predict risk factors, showcasing non-invasive risk assessment capabilities. Similarly, Siemens Healthineers and ACC's collaboration on an AI tool analysing echocardiograms, automating parameter measurement and heart function anomaly detection, facilitates early heart condition diagnosis."

Dr Apurrva Sawant, preventive healthcare expert, and founder of Careport says, "AI-powered diagnostics involve the use of machine learning algorithms to analyse large amounts of data from various sources such as medical records, genetic information, wearable devices and imaging scans. This enables the early detection of diseases by identifying patterns and anomalies that may go unnoticed by human physicians."

Another health expert, Dr Pravin Kahale, cardiologist at Kokilaben Dhirubhai Ambani Hospital says, "By analysing large volumes of data, AI provides insights into potential cardiovascular risks before symptoms even manifest. AI also analyses plaque build-up in the arteries and its potential to cause a heart attack."

Further, Arindam Sen, CEO and director, Heartnet (IoT-based solution provider) says, "The integration of AI extends to risk stratification for CVDs as well. By considering factors like age, gender, and medical history, AI-powered diagnostic tools can predict the likelihood of an individual developing a specific type of CVD. This personalised approach enables physicians to tailor treatment plans based on individual risk factors, paving the way for more effective and personalised patient care."

These advancements highlight AI's pivotal role in early CVD detection for timely interventions and improved patient outcomes.

Khare highlights incidents she witnessed first-hand where AI helped individuals seek urgent medical attention: I came across a case in which AI played a significant role in saving the life of patients and also in making accurate decisions for further course of management (with the help of AI we were able to confidently avert invasive surgeries and kept them on optimised medical treatment for years). In another incident, a patient in his late 30s came to me stating his wearable device was alarming for fast heart rate consistently.

Initially, he felt it might have been some device error but after being alarmed he chose to seek help. It was found that he had developed atrial fibrillation (an irregular, often rapid heart rate that can lead to blood clots in the heart) which was confirmed in 12 lead ECG. On taking a detailed history, he revealed he had two episodes of a sudden loss of consciousness in the past (which he confused with random tiredness). Further investigations were done and diagnosis was established. He was put on antiarrhythmic drugs and oral anticoagulants and kept on regular follow-up. This timely intervention saved him from probable future calamity.

AI aids doctors in treating CVD
In a tech-driven world where exchanging documents using phones saves us from lengthy government procedures, AI has also enabled doctors to treat patients thanks to AI-enabled mobile phones and wearable devices.

Patients send their ECGs and echocardiography reports which doctors can study on the go using their AI-empowered smart devices. This aids healthcare professionals in expediting medical attention and saves crucial time in several incidences.

AI also aids in post-treatment or post-surgery care of cardiovascular patients by analysing vast amounts of patient data, including vital signs, lab results, and imaging scans. It can identify patterns and trends that might not be immediately obvious to clinicians, helping in the early detection of potential risks such as irregularities in heart rhythms, clot formation, or other complications.

AI also streamlines the workflow for healthcare professionals by automating tasks such as image analysis and data interpretation.

AI cannot replace doctors
When asked if, in the coming future, AI has the potential to replace doctors, Khare, Sawant, Kahale and Sen say it is impossible.

"AI technology can complement doctors in some areas but cannot replace them. AI excels in tasks like analysing large volumes of data, making accurate diagnoses based on patterns and suggesting treatment plans which are originally fed into systems by humans, based on research and clinical knowledge. However, doctors offer empathy, distinct judgement, and the human touch, crucial in patient care that AI cannot fully replicate," says Khare.

Sawant says, "AI has come on leaps and bounds in healthcare, but human input and surveillance are still relied upon. Humans are unique in the sense that they can notice behavioural observations in a way that no machine can."

"While AI facilitates early detection, it is not a substitute for regular medical check-ups and diagnostic tests conducted by healthcare professionals. AI systems cannot yet replace the medical profession. They can only assist and aid in early diagnosis and treatment," opines Kahale.

Sen too states, "While AI helps with the screening process, the final decision and treatment plan needs to be evaluated and approved by a human medical professional alone."

The technology cannot be completely relied on
Our experts have listed down certain limitations to AI that prove we cannot completely rely on AI.

1. Healthcare involves complex interactions between patients and healthcare providers that go beyond data analysis. Compassion, empathy, and effective communication are essential aspects of patient care that AI cannot replicate.

2. Decision-making in healthcare often involves ethical judgements that require human values, empathy, and a deep understanding of complex situations. This calls for greater transparency about the types of data that are being shared and to whom and to clarify the purpose of it being shared.

3. AI is trained on historical data and may struggle with novel or unforeseen scenarios. Healthcare providers need to rely on evaluation and human judgement in such crucial cases.

4. The legal and regulatory landscape in healthcare involves complex ethical, legal, and social issues. Human professionals are needed to navigate these aspects.

5. No AI algorithm is foolproof and can be faulty if the ML or machine learning algorithm has not been trained properly. It has often been seen that AI algorithms have a higher sensitivity but a poor specificity which increases the risk of overdiagnosis and further evaluation.

Accessibility challenges in India and the road ahead
India is witnessing a significant surge in AI-driven healthcare applications, from diagnosis to treatment and beyond. Sawant shares, "According to recent statistics, the Indian healthcare AI market is expected to reach USD 1.6 billion by 2025. The Covid-19 pandemic accelerated the adoption of telemedicine, and AI is enhancing this shift. AI virtual health assistants can interact with patients, answer their queries, and even monitor vital signs remotely, ensuring continuous care without the need for long and arduous journeys to urban health centres."

Adding to this Kahale says, "Despite these challenges, AI technology is increasingly being used in India, and there is a growing interest in integrating it into the healthcare system. Many companies in India have started adopting AI applications in image analysis, diagnostics, and patient care. However, the pace of this adaptation depends on various factors such as regulatory approvals, technology upgrades by institutions, and collaboration between technology providers and healthcare institutions. The exact timeline for widespread adoption is difficult to predict, but it is progressing rapidly, even in India."

Yet, according to Khare, AI technology for cardiovascular diseases faces accessibility challenges in India due to limited infrastructure, high costs, insufficient skill and awareness among healthcare professionals, and regulatory hurdles.

She states, "Achieving optimal utilisation necessitates collaborative efforts between the government, private sector, and healthcare institutions. Key steps include investing in accessible technology, providing training to healthcare professionals, developing policies for affordability and inclusivity, and establishing robust regulatory frameworks. The timeline for achieving widespread use depends on financial investments, policy changes, technological advancements, and stakeholder commitment, making an exact prediction challenging. However, concerted efforts could potentially expedite progress within the next decade, fostering more comprehensive AI integration in cardiovascular care across India."