Analysts from Frost & Sullivan are convinced that artificial intelligence and machine learning technologies will become more influential in this area every year.
The same opinion is shared by Naveen Jane, founder of the startup Viome, which is engaged in the development of medical technology. In an interview, a businessman said that a real AI tsunami is coming in healthcare.
Further, in the article, we will talk about HIPAA compliant app development, projects, and working solutions of artificial intelligence in medicine.
AI algorithms are already in full use in medical practice, helping doctors identify diseases and prescribe treatment.
Radiological clinics use the Watson Health platform from IBM in the United States, India, and Thailand. A cognitive program based on it can detect cancer or heart problems in patients. The Russian development of TeleMD also diagnoses cancer and also assesses the risks of their development. DeepMind Health, Google’s technology, operates in the UK ophthalmology clinic, identifies some eye diseases, and recommends how to treat them.
Last year, information appeared that in one of the hospitals in England began testing AI, which conducts ultrasound diagnostics of pregnant women. Software called ScanNav in real-time and, in parallel with the doctor, examines the fetus for pathology.
BIDMC microbiologists have developed a smart microscope that uses AI to diagnose deadly blood infections. Its neural network has studied 100 thousand images with harmful bacteria. It is now able to sort them by visible signs with an accuracy of 93%.
Based on AI, software developers release services to monitor patients. Doctors and scientists examine the results and then conduct clinical trials.
- Duke University professors have created Autism & Beyond and mPower apps that track the symptoms of autism and Parkinson’s disease to help improve diagnosis. Later, Apple developed the Health Records API software on their basis, so that users can share data with medical researchers even through third-party applications.
- Scientists from the Massachusetts Institute of Technology, together with specialists from the Central Hospital from the same state, created an AI system for monitoring human sleep. It tracks radio signals reflected from a person, analyses the pulse, respiration rate, and can distinguish deviations from the norm. The development will help doctors remotely inspect patients’ sleep and, if necessary, adjust it.
In 2018, the American medical journal Anesthesiology published the results of a study of artificial intelligence, useful in surgical treatment methods. The article deals with a machine learning algorithm for predicting hypotension during surgery. AI analysed the data of more than a thousand patients, who spent a total of almost 10 thousand hours on the operating table. He learned to predict anomalies 15 minutes before they occurred with 84% accuracy, with the same – in 10 minutes, and from 87% – in 5 minutes.
- Qventus – a monitoring system for hospitals from the same startup. He monitors the actions of clients from recording in the registry to discharge, knows how to predict the deterioration of patients’ well-being, analysing their condition. Also, with the help of this AI, Mercy Clinic reduced the number of unnecessary tests by 40% over four months based on similar customer complaints.
- Jvion’s machine-learning solution identifies patients at risk of re-admission to the hospital within 30 days of discharge. Also, it provides recommendations on health care and disease prevention.
Pharmaceutical giants such as Sanofi or Novartis are resorting to startups developing medical innovations to search for new drugs. Biochemicals manufacturer Roche has acquired Flatiron Health, a company that uses machine learning to process data.
Since 2012, the Atomwise startup has been using neural networks to search for more effective drug formulas. AtomNet deep learning system checks 10 million chemical compounds daily, predicting which ones will interact best. A similar algorithm is used by the biopharmaceutical company Berg Health.
The compounds found can be useful in combating the cause of the disease, but this does not guarantee that the human body will respond well to them. NorthShore Medical Center, among other things, is engaged in pharmacogenomics. In essence, it studies the effect of drugs on individual people as part of the MedClueRx project. The system determines which medicines are suitable for a particular patient with epilepsy, infectious diseases, depression, gastrointestinal diseases.
The scientific journal Nature Microbiology published an article about VarQuest last year. He can detect ten times more variations of antibiotics than before it was found for all the time of similar requests.
Developers use AI technology to create a wide range of smart assistants: from personal doctors to robotic surgeons.
- Woebot is a chatbot for dealing with depressive thoughts and conditions from psychologists at Stanford University in collaboration with AI experts. It works based on cognitive-behavioral therapy, capable of changing behavioral patterns and destructive stereotypes. A similar application is developing a startup, Wyse.
- The “Mobile Clinic” DOC + from a Russian startup of the same name allows you to consult with a doctor remotely, call a specialist at home, and reserve medicines at nearby pharmacies. The application also creates an individual electronic medical record, accessible only to the user, which he can share with the doctor.
- Da Vinci is a well-known robot surgeon with artificial intelligence working in more than one hundred clinics around the world, including in Russia (hospital No. 31 in Moscow). Minimally invasive surgery is no less successful in SenEnt TransEnterix.
- And in China, scientists were the first in the world to clone pigs using artificial intelligence robots successfully. Smart micromanipulators themselves collected and transferred donor DNA to surrogates. In April 2018, two sows brought a litter of 13 healthy piglets.
“Medical futurist” Bertalan Mesco once said that artificial intelligence is a 21st-century stethoscope. He implied that at first, the medical community did not want to recognise such a simple instrument as a stethoscope. It took several decades for doctors to start using it. The same thing is happening with AI: someone uses it to the extent possible, while someone is afraid of it.
However, technologies of artificial intelligence, machine learning, and neural networks greatly simplify the lives of doctors and their wards. Innovations in medicine make it possible to diagnose diseases more accurately, find drugs faster, and monitor patients. And this is only a small part of the opportunities that AI has brought to the healthcare sector.