AI in Healthcare
It is said before the invention of the stethoscope is a pretty funny incident. In earlier days, the doctor used to place his ear to the chest of the patient to hear the heart sounds, when a female patient came to the clinic for a check-up, the male doctor was rather awkward to do so. Which, they started using a tube in order to listen to them, which led to the discovery of the stethoscope by Dr. Rene Laennec. The development of technology and the development of healthcare went hand in hand. Modern medicine would not be called ‘modern’ without the contribution of technology to it. Be it use ECG, which uses principles of electrical conductivity to know about the heart function, or Ultrasonography which uses piezoelectric crystals to convert electric energy to sound energy in order to form images of internal organs or the radiology machines which use principles of magnetism, laparoscopic and robotic surgeries, and the list goes on. In ICUs, operation theatres, and other critical centers in the hospital, there are various monitors that help to constantly monitor the vital functioning of the body. As said by the great Narayana Murthy sir, “Effective use of technology is important to deliver healthcare. By leveraging technology, you can bring down the lack of access and cost of healthcare”. With the improvement in the healthcare system due to revolutionary equipment, a lot of time is saved for the healthcare facilitators and access to healthcare has also increased to unknown bounds.
After the introduction of Artificial Intelligence in the late 1950s. The reach of artificial intelligence technology innovation continues to grow day by day, changing all industries as it evolves it has been impacting various domains like finance, marketing, the gaming industry, and even the musical arts. However, Artificial Intelligence has the largest impact in the field of Healthcare. AI can help doctors make better decisions, manage patient data information effectively, create personalized medicine plans. Clinical decision support in ai can help to assist doctors to make better decisions faster with pattern recognition of health complications that are registered much more accurately than by the human brain. Information management will be an addition for both physician and patient, with patients getting to doctors faster or not meeting him or her at all by means of telemedicine which reduces time and cost. The ability to analyze large amounts of patient data to identify treatment options. The technology can identify treatment options through cloud-based systems able to process natural language and acquire personalized medicine. According to the latest report by PwC, AI will contribute an additional $15.7 trillion to the world economy by the year 2030 and the greatest impact will be done in the field of healthcare.
Artificial Intelligence in Telemedicine
Artificial Intelligence is not only for lab work it has developed in telemedicine also. Telemedicine allows many long-distance patient and clinician contact, advice from doctors, reminders, education, intervention, monitoring, and remote admissions. As the world stopped to move during the COVID times but doctors must do their duty at those hard times too. As it was an emergency in the medical field doctors would not take minor cases unless and until it was an emergency for the patients felt by the doctors. All the surgical operation for ortho has stopped until everything was back to normal.
So, non-emergency cases or minor cases can be monitored through sensors from which information can be stored. Wearable technology is devices that are worn nearer to the skin mostly hand-designed which is used to monitor health. The recent wearable devices can monitor various factors and collect data on Heart rate, Calories burned, Steps walked, Blood pressure, Release of certain biochemicals, Time spent exercising, Seizures, and physical strain. These devices allow for constant monitoring of a patient and the ability to notice changes that may be less notified by humans. Information collected by the devices can be compared with previous data and other data using artificial intelligence and may alert the doctor in a critical situation.
“Precaution is always better than cure”, is the slogan behind the release of the Apple watch. Apple used Artificial Intelligence to build a watch that monitors an individual person’s health and fitness. This watch collects data such as a person’s heart rate, step counter, sleep cycle, activity level, breathing rate, blood pressure, etc. and keeps a record of these measures 24/7. This collected data is then processed and analyzed by using Deep Learning and Machine learning algorithms to build a model that predicts the risk of a severe heart attack. An individual named Scott Killian saved his life using his Apple watch.
To aid the patient sometimes doctors may not be available not a replacement but for aiding or being a helping hand for the doctors Chat-bot therapy can be taken as help. A chatbot is an application that was developed using a machine learning algorithm and Natural language processing. Chat-bot has been very successful in news media, social media, banking, and customer service. Patients are comfortable having a doctor one to one conversation, and artificial intelligence is what makes a chat-bot sound like a human. chatbots will not only improve care delivery, but they will also lead to significant healthcare cost savings and improved patient care outcomes soon. The medical chatbots can reduce hospital visits the lockdown time people could not visit the hospitals where this could be one of the better replacements in such cases. Reducing unnecessary treatments and procedures and decreasing hospital admissions and readmissions.
Artificial Intelligence in Medical Diagnosis
Diagnosing diseases correctly will take years of training in diagnostics often a time-consuming process. Deep learning algorithms have made advancements in automatically diagnosing diseases. Machine Learning algorithms learn to monitor or evaluate the patterns similarly doctors see them. But the difference between a doctor and an AI machine is that it needs thousands of cases to learn and good data processed data because machines cannot read between the lines. The AI models are help full in Detecting lung cancer, sudden cardiac death, or other heart diseases based on cardiac MRI images, skin lesions in the skin, and finding indicators of diabetic retinopathy in the eye. There are plenty of data available in these fields’ algorithms are becoming good at diagnostics.
Global Market Insights states that “Medical imaging and diagnosis powered by AI should witness more than 45% growth to surpass USD 3 billion by 2024.”. Artificial Intelligence is revolutionizing the image diagnosis field in medicine, with the help of Neural Networks and Deep learning models. It has taken over the complex detail analysis of MRI scans and made it a simpler process.
- MRI scans are very difficult to analyze due to the amount of information they contain. A normal MRI scan analysis takes several hours, and researchers are trying to formulate an outcome from large data sets, wait for hours for a computer to regenerate the scans.
- Large and complex data sets can be analyzed with the help of neural networks and this is exactly what a team of researchers implemented at MIT. They developed a neural network which they called “VoxelMorph” that was trained on a data set of approximately 8000 MRI scans.
- A neural network functions by inputting data at one of the ends which undergoes a transformation throughout the network until the final desired output is achieved. A neural network works on the principle of weights and bias.
VoxelMorph was successful in beating conventional MRI analysis methods. The neural networks took a few seconds to perform the MRI scan analysis, the same analysis that used to takes hours for a conventional MRI program.
Artificial Intelligence in Drug Discovery
There are dozens of pharma and health companies currently working on Artificial Intelligence to help with drug discovery and improve the lengthy recovery processes tied to inventing or discovering and taking drugs all the way to market. Drug discovery is another great place for AI to cope up with the pharma companies able to include cutting-edge technology into the expensive, difficult, and lengthy process of drug discovery.
The benefits of AI are corresponding with the focus on timesaving and pattern recognition upon testing and identification of newly developed drugs.
Drug development has three main stages. The first is to identify targets for intervention which is understanding the biological origin of the disease. Need to identify good targets for treating the disease mostly proteins. Machine learning algorithms can analyze and identify target proteins. The next stage is to Discover drug candidates in which ML can learn to predict the suitable molecular structure. Filter out the best options and have fewer side effects. Speed up clinical trials, hard to find candidates for trials. If choosing the wrong one might be living in danger. ML can again help in picking a suitable person from the trial. The algorithm can help identify patterns that separate good candidates from the bad.
In early-stage drug discovery, start-ups such as Verge Genomics or BenevolentAI are known to adopt algorithms that comb through portions of data for patterns that are too complex for humans to identify, saving both time and innovating a way that we may not have been able to do.
Insilco is another company with a heavy focus on AI, which has taken a different approach by using AI to design treatments that are not yet found in either nature of chemical libraries. An approach of using Artificial Intelligence to simulate clinical trials before human trials have also been seen with leaving plenty of scopes available for what AI can create.
Artificial Intelligence in Medical Assistance
As the need for medical assistance has increased, the development of AI-based virtual nurses has tremendously increased. According to a recent survey, AI-based Virtual nursing assistants is found out to be at the maximum near-term value of USD 20 billion by 2027.
“Sensely” is one of the examples of a virtual nurse that uses speech recognition, Natural Language Processing, Machine Learning, and wireless integration with medical devices like blood pressure cuffs to provide medical assistance to all the patients.
Some of the key features that the virtual nurse, Sensely provides:
- Self-care
- Nurse Line
- Clinical advice
- Scheduling an appointment
- ER Direction
With such revolutions in the field of medicine and healthcare, it is clearly seen that despite the risks and the so-called ‘threats’, Artificial Intelligence is benefiting us in so many ways.
Artificial Intelligence in Decision Making
Artificial Intelligence has played a very major role in decision making. Not only in the healthcare industry but also in AI has improved businesses by studying customer needs and evaluating any potential risks.
A powerful and impactful use case of Artificial Intelligence in decision making is in the use of surgical robots that can minimize the errors and variations and eventually help in increasing the efficiency of surgeons. Da Vinci is one such surgical robot, allows professional surgeons to implement complex and critical surgeries with better control and flexibility than conventional approaches.
Some of the key features of the Da Vinci include:
- Aiding and helping surgeons with an advanced set of instruments.
- Translating and converting the surgeon’s hand movements at the console in real-time.
- Producing 3D high-definition, clear, and magnified image of the surgical area.
Surgical robots not only assist in decision-making processes but also improve the overall performance by increasing accuracy and efficiency.
Knowing the current limitations of AI in healthcare
Although AI in healthcare has huge potential and very efficient, as with most developments in the technological space, there are several known current limitations in it.
· Initial adoption issues
Experiencing complex problems with the introduction of any new technology are not rare, but it must be overcome for large scale adoption of AI to occur in the healthcare market.
Ultimately, the adoption of AI will attract stakeholders and investors who will invest in AI and successful case studies need to be highlighted and then presented for future encouragement. These case studies will require some early adopters of healthcare companies to kick-start the process.
· Data privacy concerns
Privacy within healthcare is a very important factor that is extremely sensitive and thus confidential. For utmost confidence in the technology, systems should be put in such a place to ensure full data privacy and protection from hackers. But privacy concerns should not be a preventative from adopting AI in the healthcare space. In fact, last year we did a story on how AI can help healthcare data security.
· Compliance with regulations
HIPAA (Health Insurance Portability and Accountability Act) and several other patient data laws are subject to the approval of governing organizations. e.g., FDA (Food and Drug Administration) to ensure that federal standards are maintained.
The sharing of data among a variety of databases poses tough challenges to HIPAA compliance and very good care must be taken around these areas if future developments are to succeed. As companies are developing software, therefore AI, are also required to comply with Hitrust rules, current rules are regulations are known to be a barrier to AI adoption.
· Black box difficulty
AI, Deep Learning, and Machine learning do not have the ability to enquire the question ‘why?’. As a result, the logic behind decisions is not yet justified, meaning mostly guesswork is required to how the decision was made.
Why and How the decision that has been made is key to the information within the treatment plan. With a lack of reasoning, it can form a lack of confidence within the decision, potentially rendering the technology as unreliable or untrustworthy by both patients and medical professionals.
· Stakeholder complexities
When it comes to the stakeholders within the adoption of AI in healthcare, everyone, including patients, doctors, insurance companies, pharma companies, healthcare workers, etc. are key.
Resistance to pursue the technology at any of the above-mentioned levels would result in issues and potential failure to the incorporation of the technology in the macro. A stakeholder is one of the top 10 reasons why the healthcare industry is not innovating enough in 2019.
· Clinical decision support
Diagnostic errors are found out to be 60% of all medical errors and an estimated 40,000 to 80,000 deaths every year. As a result, artificial intelligence has been employed in a variety of different areas to reduce the toll and number of human errors.
There continues to be significant pushback when it comes to AI adoption in the clinical decision support process as doctors and scientists continue to approach the topic of AI with incredible caution.
· Easy to use with a clear output
With minimal operator training needed and design with common output formats that directly interface with other medical software and health record systems, the system is incredibly easy to use and simple to implement.
A clear output from the system allows a minute to identify whether the exam quality was of sufficient quality or the patient is negative for referable DR (Diabetic Retinopathy) or the patient has signs of referable DR. Following signs of referable DR, further action in the form of a human grader over-reading, teleconsultation, and/or referral to an ophthalmologist may be suggested.
Summary
AI is already helping us more efficiently in telemedicine diagnoses diseases, develop drugs, and medical assistant.
This is just the beginning of the story it is unlikely to say AI will replace doctors overnight. The more we step into digitizing and processing our medical data, the more we can use AI to help us find valuable patterns we can use to make accurate, cost-effective decisions in analytical processes. Currently, AI products or chatbots can replace doctors neither take over the care and support which a human doctor can give. Healthcare is a basic necessity that has undoubtedly improved beyond measure due to the contribution of technology and hopefully continues to be so in the upcoming generations. The best way to say this is by combining both worlds and creating a better future for healthcare.
Authors: S.Abhishek, Tejas Kachare, Om Deshpande, Rushikesh Chounde, Prachi Tapadiya.
We hope you found this blog interesting, feel free to drop your queries in the comments below. Stay tuned for more!
THANK YOU….