Spectrum of Detection, Drug Delivery, and Personalized Medicine Broadens with The Emergence of Artificial Intelligence
The Implementation of AI Technology to Speed up Detection of Sight-Related Diseases, Chemicals That Will Lead to Drug Discovery and Personalized Medicine for Patients with Multiple Myeloma Would Help Medical Professionals
The emergence of artificial intelligence (AI) in the medical field has resulted in the shift of paradigm. Research activities have made several breakthroughs. Alphabet-owned DeepMind has developed a system for rapid detection of sight-related diseases. This system is in the testing phase. Moreover, a team of researchers at the University of Glasgow has developed a system powered by the machine learning algorithm to explore chemical reactions and determine the combination of chemicals that would lead to drug delivery. This would reduce the cost of drug discovery.
In another implementation of the AI platform, researchers from the National University of Singapore (NUS) has developed an AI platform to design a personalized drug combination rapidly for patients with multiple myeloma. Implementation of AI technology has been gaining momentum in the medical sector with new discoveries and taking the global industry to the worth of few billion dollars. According to the research firm Allied Market Research, the global artificial intelligence in medicines market is expected to reach $18.11 billion by 2025. Following are some of the research activities that have shifted the paradigm in areas of medicine through the emergence of AI platforms.
DeepMind’s AI for eye diseases:
Technologically advanced solutions would help in detection of various diseases to help doctors begin diagnosis as soon as possible. DeepMind, a London-based AI firm owned by Alphabet, has planned to develop a product to help doctors in the detection of eye-related diseases through a general eye scan. According to a research published in the scientific journal Nature Medicine, the company has developed an AI software for better detection of signs of eye diseases in comparison to human doctors. It collaborated with London’s Moorfields Eye Hospital and the University College London Institute of Ophthalmology. Machine learning algorithm has been used for training AI in this system.
For now, the company tested its software on nearly 1,000 scans that have not been used for AI. This was done to set the benchmark for the system. Then the performance of scans was compared with four senior ophthalmologists and four optometrists who had a wide experience in interpreting OCT scans. The researchers discovered, for nearly 50 eye diseases, the AI system was able to take the correct referral decision with 94 percent accuracy, which is better than most of the humans. This system is among the initiatives of Alphabet to improve healthcare with the help of AI. They plan to launch clinical trials of the technology in 2019. Following the successful trials, the regulator-approved product will be created. It will be offered free of cost for the first five years.
AI to ease up drug discovery:
AI can be helpful in discovering drugs, thanks to machine learning algorithms. Researchers have trained robots to inspect a huge number of chemical reactions to reduce the cost of discovery of new molecules for drugs. They also believe that the “self-driving” system would help in finding new chemical products such as polymers, materials, and molecules for various applications. According to a research published in the journal Nature, this approach would help in discovering new molecules and reactions rather than confining to general rules of organic synthesis and known database.
“This approach is a key step in the digitization of chemistry, and will allow real-time searching of chemical space leading to new discoveries of drugs, interesting molecules with valuable applications, and cutting cost, time, and crucially improving safety, reducing waste, and helping chemistry enter a new digital era,” said lead researcher Leroy (Lee) Cronin, Professor at University of Glasgow in Britain. The team of researchers found nearly 1,000 reactions with the help of a combination of 18 different starting chemicals. After exploring nearly 10 percent of possible reactions, the system was able to determine combinations of starting chemicals to form new reactions and molecules with over 80 percent accuracy.
Personalized drug combinations for myeloma patients using AI:
The paradigm of combination therapy design would change with the emergence of new technology. A team of researchers from the National University of Singapore (NUS) has developed an AI technology platform to design a personalized drug combination rapidly. This platform is designed for patients with multiple myeloma, which is a blood cancer type. The AI platform would help in designing an effective personalized drug combination and determine patients who will be responsive to treatments within a week. Bortezomib-containing drug combinations have been used currently as the first- and second-line treatment for multiple myeloma. Many patients have become resistant to these treatments and new combinations are required to be found.
Researchers have developed Quadratic Phenotypic Optimization Platform (QPOP), an AI platform to accelerate drug combination design and determine the most effective drug combinations for individual patients with the help of small experimental data sets. Using a few drops of blood or bone marrow sample, this AI platform would determine the drug response on the cancer cells of a specific patient. The optimization and personalization capacity of drug combinations with high speed helps in improvement of accessibility to personalized medicine for patients.
The spectrum of detection of diseases, drug delivery, and personalized medicine has widened with the emergence of new AI technologies. The recent research and development activities would find more applications in medicines and help doctors speed up decision making and treatment procedures.