Introduction
According to Accenture, artificial intelligence (AI) in the healthcare industry could potentially save $150 billion for the US annually.
As such, machine learning and deep learning algorithms from AI are being rapidly deployed for medical diagnostics, doctor consultation, personalized disease treatments, electronic health records, drug discovery and much more. Amazon’s Alexa provides medical advice for users’ symptoms and can manage blood sugar data for diabetic patients. Google recently released an open-source version of DeepVariant, an AI genomics tool, on its cloud platform to increase the accuracy of sequencing a person’s genome, reducing error rates by 50%.
Beyond these gigantic tech companies, this industry landscape examines over 200 companies (predominantly startups) utilizing AI in conjunction with doctors and researchers to tackle some of the most pressing health dilemmas.
Methodology
The 4 main categories of the landscape represent the stakeholders in healthcare who are impacted by AI: patient, doctors, researchers, or telehealth (which is the interdependence between the former two).
Within the 4 categories, the 16 subcategories sort the tech companies most relevant to patients’ specific needs, doctors’ workflows, researchers’ methodology, and interactions between patient and doctor. Sources to identify and aggregate companies were TechCrunch, AngelList, CB Insights, Redox Engine, Nanalyze, and various medical blogs of company websites. Crunchbase provided company information after categorizing the list of startups.
Key Findings
- The United States accounts for about 70% of all AI healthcare startups with the remaining 30% from European regions, Israel, Asia and Canada.
Some companies specialize in subcategories. For example, Israeli companies, Zebra Medical Vision, AiDoc, DiA Imaging Analysis, and MedyMatch (“Medical Imaging” and “Hospital”) incorporate AI algorithms that enable radiologists to spot abnormalities, naked to the human eye for a more rapid diagnosis in medical imaging scans.
United Kingdom companies Benevolent AI, Exscientia, and healx (“Drug Discovery”) apply AI to aid scientists in identifying novel drugs for selected diseases. Complementary to the “Drug Discovery” companies, Antidote another United Kingdom company, is expediting clinical trials (“Information and Clinical Trials”) with its AI-driven platform to match patients and researchers.
2. After a period of steady growth, the number of new startups nearly doubled from 2013 to 2014 with a slightly negative drop in 2015.
For 2014, all 4 main categories had a positive net increase in startups. Patient-facing companies experienced the most growth, tripling from 4 to 12 companies. Half of the newly-created companies (DietSensor, BioSerenity, Aira, and CliniCloud) were wearables and devices. In 2014, the wearables market was poised to grow by 350% (CNBC) due in part to Fitbit’s release of the Flex and Force bands in May and October of 2013, respectively. 3 other companies (Vitagene, Arivale, Suggestic) were based on personalized genetics. Concurrently, one of Healthcare IT News’s 2014 trend was expedienting DNA sequencing for personalized medicine.
In 2015, there was a wave of 18 new doctor-facing companies, resulting in a positive net increase of 6 companies. This balanced out the net decrease in the other main categories for a minimal decline in 2015 in terms of total new companies. The new doctor-facing companies were evenly distributed amongst the subcategories in medical records, data analytics, and medical imaging.
3. For cancer and cardiology, companies are exploiting AI in all of the 4 main categories.
The data above is limited to 63 companies, and excludes those focused on general wellness and medical adherence, as well as those that did not specify a disease. 6 companies targeted more than one type of disease and were not included in the 63 companies to avoid double counting. (These were the patient-facing companies Magnea, Clear Genetics, elyse28, Arivale and the research companies Recursion Pharmaceuticals and Pathway Genomics).
Given the various types of cancers and its bleak prognosis as the second leading cause of death in the United States, cancer is being addressed by:
a. Patient-facing companies-The subcategory, “Skin” provides medical information on skin cancer, while Cyrcadia Health in “Women’s Health” offers a wearable to monitor for breast cancer.
b. Telehealth companies-In “Disease Management”, Tempus creates personalized cancer treatments with physicians and Belong is a personalized solution for families to fight cancer.
c. Doctor-facing companies-Flatiron Health and Oncora Medical are leveraging data analytics for better patient care in clinical communities. Infervision, CureMetrix, and Koios Medical, are targeting lung and breast cancer with AI-driven image recognition and diagnostic tools.
d. Research-Genomics company Molecular Match provides a dynamic molecular search engine to connect users to the latest cancer treatments. Recursion Pharmaceuticals and Insilico Medicine from “Drug Discovery” explore new drug treatments for scientists through biological image datasets and custom drug discovery engines.
Compared to cancer, solutions for diabetes are mostly wearables and devices from patient-facing companies (Sano, DietSensor, Magnea). This originates in part from expanding the capabilities of the Google Glass to create effective, remote glucose monitoring solutions.
Final Thoughts
AI is empowering patients with their health data, particularly through chatbots and wearables & devices that enable more individualized healthcare. Data collection from patients is increasing the doctors’ database of medical intelligence and researchers’ practices for higher accuracy and efficiency in novel treatments.
This industry landscape is one comprehensive snapshot, but fluid in that companies can cater to multiple stakeholders (ie. Ada Health, a telehealth company also partners with researchers for diagnostic intelligence) and incorporate features that are reflective of other sub-categories (ie. Lark from “Disease Management” pulls data from Apple watches and smartphone data).
Overall, the healthcare industry is becoming more horizontal across stakeholders, diseases, and technology. At the forefront, AI is an innovative means to better serve the health needs of individuals and communities.
