The challenges of clinical impact with artificial intelligence – Prasanth Madavana

Artificial intelligence is not magic. AI tries to understand patterns from data much more efficiently, a quick process that’s a limitation for humans. Then it creates models for future predictions and inferences.

Artificial intelligence in healthcare

The story is strong on how the human brain and expertise are the life and soul of all possible innovations and revolutions. Whether it’s the 7 Wonders of the World, or the invention of the steam engine and airplanes, ideas generated from the threads of the human brain have always made the way of life easier. Today, technology has been massively accelerated in the healthcare industry. With the current pandemic, use is rapid and the impact is massive. From a medical chatbot to the detection of critical diseases like breast cancer, the industry is booming with the benefits of artificial intelligence. A person staying in a rural area in India is now aware of the video consultation and assesses the risk of covid by answering some poll questions in the vernacular. When humans realize that data is the secret sauce for creating possible innovations, there is no looking back. According to Amazon CEO Jeff Bezos – “I think healthcare is going to be one of those industries that is uplifted and improved by machine learning and artificial intelligence.” The intensity of the COVID19 pandemic would have been much greater if there had been no contact tracing process. AI / ML applications have proven to be wonders for assessing an individual’s risk of covid and for strategically planning the vaccination campaign.

Let’s take a moment and realize how completely our daily life has been dependent on artificial intelligence. From ordering food from our favorite food chains to getting daily recommendations on OTT channels, AI has managed to disrupt almost every industry. But as far as the healthcare industry is concerned, the impact is not yet fully realized. There are few challenges to achieve meaningful clinical impact with the use of AI.

Challenges

The healthcare industry is one of those arenas where human interaction and empathy are required. Artificial intelligence is not magic. AI tries to understand patterns from data much more efficiently, a quick process that’s a limitation for humans. Then it creates models for future predictions and inferences. The power of reasoning, learning new patterns outside of past data, and demonstrating empathy have yet to be realized in this technology. As Nick Bilton, a leading technology columnist for the New York Times, said, “Artificial intelligence upheavals can escalate quickly and become more frightening, even cataclysmic. Imagine how a medical robot, originally programmed to rid cancer, could conclude that the best way to erase cancer is to exterminate humans who are genetically predisposed to the disease. With reasoning features, these problems can be eliminated. The potential of BIG DATA is enormous. The data vary from individual to individual and from population to population. If an AI model learns and adapts to the patterns of a particular sample, there will be issues of underfitting, overfitting, and curvefitting. The template will not work for a different set of samples. Again, the model robustness process is extremely time consuming. There are models that can detect heart rate, blood pressure, oxygen saturation, and stress level through certain features present on the face. But due to the varied facial characteristics of individuals, such as color, facial puffiness, and placement of organs on the face, the model will not produce an accurate clinical impact if robustness is not performed. When dealing with models related to human body, the accuracy of the results should be very high. Therefore, problems with model performance and model diversification over a diverse population affect accuracy. Privacy, security and data breach issues have also given rise to many ethical concerns in the successful conduct of research activities.

Barriers to adoption

The era of the Covid-19 pandemic has seen the maximum adoption of technology in the healthcare sector. It is true that it is booming. But what will happen in the post-covid era? Will there be a relapse and will patients reason by visiting a doctor in person rather than going for a video consultation? There has always been a challenge in adopting new technologies in the healthcare sector. Due to the Hippocratic Oath, doctors respect the words “Do no harm first”. Whether it’s a small mistake in the processing module, accepting ethical standards, or embracing new technologies like AI, health officials are not ready to compromise with the health of their patients. On the other hand, patients are more inclined to seek the advice of a specialist rather than a machine. Human interaction is the bread and soul of a medical consultation. Patients look for calming words, empathy, and a touch of healing after a diagnosis rather than talking to a chatbot. Yes, it is quite true that telemedicine and virtual medical consultation are currently on the increase. But it won’t generate much value for the diagnosis and treatment of life-threatening chronic diseases that require a full health investigation.

Way forward

Artificial intelligence has indeed shown a new image of the health industry and it will remain so. So, the big question is, with so many challenges, how can we leverage AI to drive maximum clinical impact in the healthcare industry? Artificial intelligence should be used to augment the 3 pillars of health service delivery – quality, cost and access rather than replacing human expertise. A doctor will easily adopt a model that can reduce burnout and make diagnosis faster with maximum precision. Additionally, adoption can be increased by adding courses on the usefulness of AI in healthcare to academic programs at medical institutions. At the same time, a strong regulatory policy specifying the requirements for data collection and AI-based models is necessary to avoid any dangerous research activity and any unethical product / service. Finally, the best way to maximize clinical impact by creating a harmonious ecosystem between the IT industry and healthcare professionals. Strong involvement of physicians in creating AI models will revolutionize the industry faster with disruptive ideas and innovations.

Source link

About Roberto Frank

Check Also

Apple is testing ways to use AirPods as a health device

Image by Natasha Connell from Unsplash When you think about how intricate the smartphone has …

Leave a Reply

Your email address will not be published. Required fields are marked *