By Srivallabh Sabnivisu
Artificial Intelligence (AI) is bringing a paradigm shift to healthcare and has become an essential part of our daily lives. The Healthcare sector is one of the most challenging sectors to be digitalized due to the loads of information needed to train the algorithms. In this process, the smartphone has become the hub of digitalization in the new era of medicine.
The most important advancements of AI in healthcare include diagnosis of diseases as well as management of chronic conditions, palliative care, predicting post-surgery complications, virtual care, or remote monitoring through a medical assistant, to name a few. To achieve these, retrospective data is collected and a hypothesis is generated to test it potentially, which is called machine learning.
Applications
In chronic conditions like hypertension and diabetes, management of these conditions play a crucial role in preventing damages to vital organs in the body. For example, uncontrolled diabetes can lead to diabetic retinopathy, and uncontrolled hypertension could lead to kidney disease. It is also now possible to predict the possibility of a person suffering from cancer in the latter part of their life. Advanced diabetes management is achieved through AI in the form of smartphone apps and glucose sensors. Glucose sensors count the amount of sugar taken by the individual, and suggest appropriate diet and exercise. In the same way, blood pressure readings can be taken by smartwatches and smartphone apps to alert the patient, as well as count steps and suggest a healthy diet to promote a balanced control of their health. Chatbots are also used to answer common health questions of customers as well as to measure their satisfaction.
Obesity has become one of the major problems of the modern world. Lifestyle changes with diet and exercise can help in efficient weight loss. A hybrid coach – AI along with human coaching – will turn out to be the best strategy for weight loss since this will provide a personalized diet plan. There is also a genetic aspect of obesity. Interestingly, it is also possible to scan the genes for any mutations and alter the diet to reduce the symptoms, and thereby its effects on other organs.
Palliative medicine is one of the many medical fields that is leveraging AI to improve the quality of care for terminally-ill patients, and also predicting the life expectancy of these patients. This information can help in determining the time needed for their care.
Administration work at hospitals, like scheduling appointments, nurse call lights, and billing systems are dealt with easily without much manpower. The machine learning systems can also completely scan the patient record and provide recommendations to the doctor, as well as rapidly analyze medical images and scans for accurate decision making.
In the process of the ‘democratization’ of healthcare, the power has slowly shifted from the physician to the patient. Nowadays patients have access to their Electronic Health Records (EHR) on their phones. They can use them to design their optimal diet or improve their physical and mental health. They can also get automatic suggestions based on their recent blood values.
There was a general tendency to avoid hospitals because of the possibility of sepsis or hospital-acquired infections. This can be minimized by tracking the hand hygiene of clinicians and surgeons using video footage and depth sensors, which use infrared light.
AI also helps in the speedy diagnosis of many diseases like brain stroke through small robotic devices. Paramedics use these devices so that the type of stroke can be alerted to the receiving hospital for them to start treatment immediately. This can reduce the toll of brain damage in the future
Benefits of AI in the healthcare domain
After a detailed discussion about the applications of AI in healthcare, the following benefits can be summarized:
· Better patient outcomes
· Simplified healthcare with better data-driven decisions
· Reduced healthcare costs
· Screen diagnostic reports on par with radiologists
· Drastic reduction in re-admission rates
· Enhanced primary care
· Reduced mortality rates
· Delivery of customized treatment plans.
Limitations
· Use of AI analytics by insurance companies might increase the rates of coverage for patients.
· The quality of input data is a critical factor in the functioning of AI assistants.
· Coaching a deep learning model on individual health has its complexities. It involves combining all the known data along with new data which can be too much intruding into the personal life.
· To train a virtual medical coach to be able to interact with a large population, all the biomedical literature needs to be continually ingested. This is a limitation considering the huge volumes of data that need to be used for the training.
· The imminent risk in the current era is cyber theft. The privacy of medical data needs to be taken very seriously as data in the wrong hands could lead to unwanted outcomes.
· As patients undergo different tests from time to time, physicians need to face the problem of handling ever-increasing loads of complex information.
AI in medicine and healthcare is still in its nascent stages. Better patient outcomes and reduced mortality with simplified healthcare and lower costs are the major advantages of AI. Right from the discovery of a drug until the time it reaches the patient, AI plays a major role in many aspects.
However, as discussed above, there are several limitations to using AI in healthcare. Nonetheless, scientists are striving hard to minimise these limitations and make them a boon to mankind.
The innovations that AI brings into healthcare systems can change things: new ways to treat people and a way – not to replace doctors – but to reduce the potential risk of mistakes happening during the process.
Sources:
· ‘The Patient will see you now- The future of medicine is in your hands’ by Eric Topol
· ‘The Digital Doctor’ by Robert Wachter
· ‘Deep Medicine- How artificial intelligence can make healthcare human again’ by Eric Topol