AI in healthcare

Artificial Intelligence is a rapidly growing sector estimated to reach a value of 150 billion USD by 2026. It has positively impacted the growth of every major industry, particularly healthcare. The applications of AI and its subfields of Machine Learning and Natural Language Processing offer a great system of support to doctors, patients, and hospital administrators by taking over menial, time-consuming tasks to simplify healthcare processes. 

Dr. Paul Weber, associate dean for continuing medical education at Rutgers’s Robert Wood Johnson and New Jersey medical schools, in a presentation on the present and future clinical applications of AI, brought attention to several areas where he has observed AI impact the field of healthcare, notably diagnosis and treatment recommendations, patient communication and care coordination. 

“We’re able to train machines to exhibit human-like intelligence and apply that in a clinical setting. We haven’t achieved human intelligence, but we’re getting close to it,” he said.

Applications of AI today

The potential for AI to transform healthcare has generated much interest in recent years. The 2018 meeting of the World Medical Innovation Forum (WMIF) was held on the topic of AI opportunities and advancements for healthcare. Leading experts in AI and representatives of IT, Pharmaceutical, government, and healthcare investment communities congregated to explore avenues in patient care where AI can make a positive impact.

AI tools are already making headway in healthcare, and various new tools awaiting approval are expected to contribute greatly towards saving lives and boosting the industry.

Technologies like pattern recognition, NLP branches such as speech recognition and translation, robotics, and Machine Learning are being leveraged to build new tools specifically targeted at healthcare. The ability of machines to perform intelligent tasks like prediction, understanding, learning, and decision-making has found successful applications in healthcare, be it genetic coding or robotic surgery.

The success of AI and ML in healthcare is dependent on access to appropriate data.  In healthcare, the most common application of traditional machine learning is precision medicine – predicting what treatment protocols are likely to succeed on a patient based on various patient attributes and the treatment context. 

Neural networks have been used for categorization purposes, say to predict whether a patient will acquire a certain disease. Deep learning, a subfield of ML, is used in radiology, for recognition of potentially cancerous lesions in radiology images. Deep learning is increasingly being applied to radionics, where radiographic images are analyzed using data characterization algorithms to discover features that escape the human eye. The combined power of DL and radionics greatly benefits the field of oncology for cancer detection and diagnosis.

NLP is dominantly applied in healthcare through applications that involve the creation, understanding, and classification of clinical documentation and published research. NLP systems build reports by studying patient notes, transcription of patient interactions, and patient engagement using AI-assisted chatbots.

Robotics is commonly used in surgical procedures in gynecology, oncology, and head and neck surgery.

What the future of AI-powered healthcare looks like

With AI emerging as a game-changer in the industry, there is a growing need for healthcare professionals and industry stakeholders to set about regulating AI development and usage in healthcare. The FDA introduced a new framework last year that enables it to pre-approve the manufacturing of adaptive AI-powered software, a move that will lead to a faster transformation of healthcare feels Dr. Weber.

Regulatory institutions must also take into account the protection of patient privacy and security. In the coming years, healthcare is expected to witness the deployment of more and more AI devices, but at the end of the day, AI will never replace the human element — the doctor-patient relationship.

15 examples of AI in healthcare you should know about

Diagnostics 

PathAI — Cambridge, Massachusetts

PathAI uses ML for increased accuracy in cancer diagnoses. It works with drug developers like Bristol-Myers Squibb and the Bill & Melinda Gates Foundation to expand its reach in healthcare.

Enlitic — San Francisco, California

Named by MIT as the 5th smartest AI company in the world, Enlitic uses deep learning for learning the needs of patients by analyzing their unstructured medical data. 

Beth Israel Deaconess Medical Center — Boston, Massachusetts

The teaching hospital is using AI for the early detection of potentially life-threatening blood diseases. Their AI machines can discover the presence of harmful bacteria in blood with 95% accuracy.

New medicine development

Berg Health — Framingham, Massachusetts

BERG used AI to research Parkinson’s Disease. It is a biotech organization that works with AI to develop medicines to treat rare conditions.

Atomwise — San Francisco, California

Atomwise’s AtomNet is a neural network used for clinical trials and can process enormous amounts of genetic compounds in a day, ensuring faster results and helping the group’s endeavour in tackling diseases like Ebola.

Benevolentai — London, England

This group is working to provide timely treatment to patients using AI and deep learning for target selection. 

Patient experience

Olive — Columbus, Ohio

Olive provides AI tools that can be used in integration with the existing software of hospitals to automate repetitive tasks like eligibility tests, saving time and money while freeing up administrative staff to focus on providing faster and more efficient patient care.

Babylon Health — New York, New York

Babylon provides AI-powered personalized healthcare services. Their chatbot connects the patient with the right doctor for a virtual or real checkup.

Cloudmedx — San Francisco, California

Cloudmedx’s technology applies machine learning to patient data to understand and provide insights for improving the patient experience. The technology can be used by hospitals and clinics to assist patients better.

Medical data mining and management

Tempus — Chicago, Illinois

Tempus works with the largest collection of clinical and molecular data to personalize and improve healthcare treatments. Its main focus is cancer research and cure.

Proscia — Philadelphia, Pennsylvania

Proscia is a digital pathology platform using AI in cancer research. It recently raised $8.3M funding for expansion.

IBM — Armonk, NY

IBM’s Watson offers a range of services to hospitals, from data management to diagnostics using AI.

Robotic surgery

Accuray — Sunnyvale, California

Accuray’s CyberKnife robotic systems equipped with 6D motion-sensing technology are being used for high-precision cancerous tumor surgeries.

Intuitive — San Francisco, California

Intuitive’s da Vinci platforms are pioneers in robotic surgery and have assisted in over 5 million operations.

Carnegie Mellon University — Pittsburgh, Pennsylvania

Their robotics department has developed a tiny, self-navigating robot called Heartlander, to assist in cardiac therapy.

Conclusion

AI capabilities in decision-making, risk management, and early diagnosis prove a highly promising investment avenue from the healthcare perspective. 

With a range of tools aimed at better diagnostics and enhanced personalized patient services, AI signals a new era of medicine underlined by quality and innovation.