Transforming Healthcare With Tech: How AI Can Disrupt Healthcare Industry

We're seeing the healthcare industry move towards Healthcare 2.0 – a digital technology-enabled and collaborative model that puts patients at the heart of care. 

There has been a steady uptick in venture capital investments in Healthcare AI over the last five years. ( Image Source :Getty )

In 2020, the covid-induced pandemic catalysed massive digital transformation in the healthcare industry globally. Almost overnight, hospitals had to virtualise triage and care through telehealth services, accepting only the most critical cases into their hospitals. The locus of care moved from physical hospitals into patients’ homes. That change has triggered a mindset shift in a very traditional industry that has until now only selectively adopted artificial intelligence (AI) and analytics in certain areas, such as radiology. 

As the global innovation hub for some of the largest health systems in the US, we are seeing the healthcare industry move towards Healthcare 2.0 – a digital technology-enabled and collaborative model that puts patients at the heart of care. 

AI and analytics play a key role in this shift – not only to transform care, but also to accelerate life-saving clinical research, transform the way hospitals operate to combat rising inflation, simplify the lives of healthcare workers dealing with burnout, enhance patient experience in a world of digital consumption, and help deliver more accessible and equitable care to entire communities through collaborative data platforms. 

There has been a steady uptick in venture capital investments in Healthcare AI over the last five years (25 per cent CAGR). Over 3,000 funded digital health startups globally are developing AI-based solutions for healthcare problems. Med Tech enterprises are a large sponsor for FDA-approved AI-ML-enabled devices. While large-scale adoption is yet to happen, experiments with AI and analytics in healthcare are proving that they have the potential to transform both the healthcare business and delivery and clinical practice in many ways.

Quicker Diagnosis, Early Detection, Improved Access

AI deployment in the field of radiology is probably the most mature. AI software enables automated interpretation of radiology images (X-ray, Ultrasound, CT, MRI etc.), improving the speed of diagnosis and treatment [e.g. Qure.ai]. Deep learning software is available to improve MRI/PET image quality and reduce image noise, reduce the time taken for imaging, reduce patient wait times and increase scanner capacity, e.g. Subtle Medical. SigTuple has built a platform for AI-assisted digital microscopy which uses robotics and AI to digitize biological samples, such as blood and urine on a glass slide to enable AI-aided remote reviews and reduce the burden on pathologists for routine microscopy work. 

Similarly, Niramai’s AI-powered platform allows early detection of breast cancer using thermography vs standard mammography. AI is also being used in cardiology as a screening tool for left ventricular dysfunction in people without noticeable symptoms, and for non-invasive diagnosis of coronary artery disease. AI-guided ECG is available to detect atrial fibrillation before any symptoms are evident e.g. at Mayo Clinic.  FDA- approved AI-powered imaging software can now detect early signs of diabetic retinopathy, a leading cause of blindness in diabetic patients. Given these early detection capabilities, AI can play a large role in monitoring patients and preventive care. AI-powered devices and sensors can monitor patient vital signs and alert healthcare professionals to potential health issues in real time.

Clinical Decision Support And Personalized Treatment

AI can improve treatment pathway design and helps genomic studies and personalized medicine. e.g. at Providence, our teams are working with clinicians and researchers to build a virtual molecular tumour board that combines genomics data, pathology data, patient history and demographic data to assist physicians with cancer diagnosis and treatment. 

AI can help draft personalized radiation treatment plans and design personalized cancer treatment regimens. Mindmaze uses machine learning to optimize rehabilitation activities for stroke patients. Ginger.io uses machine learning to recommend the best time to take medication based on each patient’s metabolism and other factors. With personalized treatments, the efficacy is greater with fewer side effects, thus having a direct positive impact on patient outcomes. 

AI also helps with the remote monitoring of patients and this can transform the way we monitor and manage chronic diseases. 

Healthcare Business And Operations

In addition to improving the clinical side, AI is helping hospitals drive efficiencies in operations and core business functions. AI-based tools are helping hospitals reduce the burden on hospital staff as well as patient wait times by automating administrative tasks such as patient registration, appointment scheduling, billing and answering standard patient questions. 

Nuance Communications, an AI-enabled "ambient clinical listening” platform helps document patient visits by accurately capturing and contextualizing every word of the doctor-patient conversation and recording it as notes. 

Hospitals are using AI-based platforms like Qventus to automate healthcare operations, accelerate patient flow, reduce length of stay, optimize inpatient capacity as well as the utilization of operation theatres. AI algorithms can also help hospitals predict patient readmissions, identify high-risk patients, forecast demand for services and enable predictive staffing. Once hospitals can anticipate demand, they can optimally allocate staff instead of being surprised by a surge in care demand. 

ML models built to predict patient “no-shows” to doctors’ appointments are also helping health systems plan operations better. With the current global shortage of healthcare workers, AI could help ease the talent crisis and burnout in healthcare through AI and automation.

Providence teams have built ML models to predict cash flow cycles, automate cash collection, and better understand likelihood of payment. The use of AI in supply chain design and management is not widespread but some hospitals and health systems have started leveraging AI to improve the efficiency, agility and reduce the cost of healthcare supply chains. AI can help in demand sensing and trigger alerts to deploy inventory, manage logistics, show product cost comparisons, and handle invoices and payments. 

Patient Experience

Embedding an AI assistant in the care pathway can improve patient experience by helping them navigate the hospital more easily. Virtual agents could automate routine patient interactions. For example, Grace, a conversational AI platform for patient engagement at Providence, is a knowledge-based decision support system that helps patients find the right care, at the right place and at the right time through a chat or voice conversation. 

It enables patients to book appointments with their providers, conduct on-demand symptom checks and receive differential diagnoses, and navigate to the appropriate type of care based upon their symptoms. Generative AI such as ChatGPT can also be leveraged to enhance this. 

A layer of generative AI could allow patients to have sustained engagement with the hospitals by answering everyday queries around their care, medication, or clarifications addressed in real time. 

Population and Community Health

In the United States, hospitals are also collaborating to build shared data platforms, such as Truveta which allows hospitals to pool their respective and vast de-identified data for Truveta to develop critical insights on population health and disease management. 

Collaboration in healthcare is likely to be a growing trend, as the past two years have demonstrated the power of data-sharing to combat the pandemic with vaccines, and the successes of this period are encouraging scientists and healthcare providers to get ahead of any future infectious disease treatments. 

The findings of such vast pools of data can benefit millions of people across communities, geographies and continents. 

For example, hospitals can also better predict resourcing based on past historical data analysis of seasonal infectious outbreaks, and community health needs (e.g., high mortality rates, diabetes, asthma, antenatal care). 

At the local government level planning can be done to allocate resources based on community-level healthcare needs, be it for education, prevention, or therapeutic care. And globally, international and national governments have the potential to develop proactive crisis management protocols for future pandemics.  

(The author is the director of strategy and operations at Providence India, a Hyderabad-based global healthcare IT company)

Disclaimer: The opinions, beliefs, and views expressed by the various authors and forum participants on this website are personal and do not reflect the opinions, beliefs, and views of ABP Network Pvt. Ltd.