In today's world of seamless interconnectivity, AI's ability to identify disease patterns through machine learning is empowering public health practitioners and policymakers with enormous potentials. This offers new hope in effectively preventing various diseases by recognizing early subtle signs of the disease and preventing it from progressing. Focused, context-specific preventive steps promote cost-savings on therapeutic care, expand access to health information and services, and enhance individual responsibility for their health and well-being to combat the rise of infectious disease epidemics.
Many startups are providing services for the early diagnosis of many diseases. An example is Onward Health, which uses predictive analytics and machine learning to build a portfolio of diagnostic tools in classifiers and analytical tools. These tools help pathologists diagnose even the most subtle change, providing more in-depth, more accurate insights from available histopathological samples. Also, Onward Health is leveraging computer vision techniques and ML algorithms to offer tools in computational pathology and mammography.
A foundational AI technology called transport-based morphometry (TBM) helps doctors identify diseases that are otherwise imperceptible to the human eye. Research shows that even before apparent signs of an illness can be seen on medical images, this technology predicts osteoarthritis conditions using machine learning to link these patterns to future osteoarthritis symptoms. There is substantial diagnostic potential for this technology as TBM can hasten disease detection, which is beneficial to many patients as they can take charge of their health early before troubling symptoms develop.
There are many startups in this field. For example, diagnostic tests and blood work to test for cancer is done by Freenome using AI. By deploying AI at general screenings, Freenome aims to detect cancer in its earliest stages and subsequently develop new treatments. Harvard University's teaching hospital, Beth Israel Deaconess Medical Center, uses artificial intelligence to diagnose potentially deadly blood diseases at a very early stage. Doctors are using AI-enhanced microscopes to scan for harmful bacterias (like E. coli and staphylococcus) in blood samples faster than is possible using manual scanning with 95% accuracy.
Another example is the Nevada-based startup Cyrcadia that has developed a breast patch to detect temperature changes in breast tissue, and the data is analyzed using machine learning algorithms. Pathway Genomics has developed a blood test kit called 'CencerIntercept Detect,' which collects blood samples from high-risk individuals who have never been diagnosed with cancer as part of a research study to determine if early detection is possible.
Similarly, New Hampshire based Breast Health Solutions applies deep learning algorithms to 2D mammography, 3D mammography (digital breast tomosynthesis or DBT), and breast density assessment. It's ProFound AI technology became the first artificial intelligence solution for 3D mammography approved by the FDA. Transpara by ScreenPoint Medical trained on over a million mammograms helps radiologists analyze both 2D and 3D mammograms. The answer is already in use in 15 countries, including the USA, France, and Turkey. Electronic health records give a lot of useful information.
Israel based Medial EarlySign is leveraging artificial intelligence to mine this EHR data to detect colorectal cancer risk much earlier. The University of Oxford has validated a study done in Israel on a population of 3 million individuals. They also have two products that identify prediabetes patients and those Type 2 patients who are at risk of developing chronic kidney disease within three years. India-based AADAR operates in the "Ayurveda-inspired" preventive healthcare space. It offers herb-based products to curb lifestyle ailments like protein deficiencies, blood sugar, indigestion, cholesterol, and obesity. Yet another startup is Prognos. It is a New York-based startup that works with the primary goal to eliminate various diseases by using artificial intelligence to predict disease and drive decisions earlier in healthcare. The Prognos Registry includes more than 15 billion medical records, and it analyzes clinical lab and diagnostics results to make predictions about an individual's risk for having asthma, lung cancer, and many rare diseases.
Podimetrics is a care management company with the leading solution to help prevent diabetic foot ulcers. Likewise, Siren Care creates Neurofabrics; the first product is Siren Diabetic Socks, which allows people living with diabetes to avoid amputations. Russian based Brain Beat Ltd develops high-tech biomedical equipment that non-invasively monitors blood glucose levels. This helps keep the disease under control and improve the quality of life of patients with diabetes mellitus.
Anticipating heart failure with machine learning
Excess fluid in the lungs often presents a diagnostic dilemma as it dictates the doctor's course of action. Clinicians rely on subtle features in X-rays that sometimes lead to inconsistent diagnoses and treatment plans. Recently, a new machine learning algorithm was developed by researchers at MIT that can look at an X-ray to quantify how severe the edema (fluid collection). The system determined the right level more than half of the time and correctly diagnosed the most severe cases 90 percent. Better edema diagnosis helps in the case of acute heart issues, but other conditions like sepsis and kidney failure.
Take a selfie to detect heart disease
A new study published in the European Heart Journal found that sending a photo selfie to the doctor could be a cheap and simple way of detecting heart disease using AI.
This study is a first of its kind that uses a deep learning computer algorithm to detect coronary artery disease (CAD) by analyzing four photographs of a person's face. Although the algorithm needs to be further refined and evaluated in more comprehensive groups of individuals from various ethnic backgrounds, the researchers believe it has the potential to be used as a screening method that could detect possible heart disease in individuals in the general population or in high-risk groups that could be recommended for further clinical testing.
Know the weather to prevent diseases
There is a significant level of evidence that the weather can influence the emergence and transmission of disease. xtLytics, LLC has pre-built predictive deep learning AI algorithms and services that enable government organizations, pharmacy, insurance, and pharmaceutical companies in engaging patient and making a better decision using multiple datasets like weather-related parameters, socio-demographic parameters, geo-location parameters, treatment/medicine-utilization pattern or claims, and patient social interaction to predict the potential number of incident cases by postal code at least 15 days in advance. This model has shown promising results in predicting Vector-Borne Diseases (Dengue and Malaria), Airborne Disease (Influenza), Atopic Triad (Atopic Dermatitis-Allergic Rhinitis-Asthma/COPD), and certain types of precancerous skin conditions like actinic keratosis and skin cancers like Melanoma, which is positively correlated with temperatures and ultraviolet radiation (UVR).
What's the Future?
AI can predict the possible turn of events using particular segments, disease susceptibilities, and previous diseases. Along with the prevention of diseases with early diagnosis, it can also map potential victims in the wake of a disaster. For example, information about respiratory diseases in older adults can be analyzed and used to predict susceptibility to infections such as COVID-19. Similarly, military veterans who have been exposed to asbestos become incredibly susceptible to respiratory infections such as coronavirus. Such kind of information is vital in controlling COVID-19 from becoming fatal globally.
Although these prospects look attractive, before we make a final decision about the applicability of AI in disease prevention, all stakeholders should insist on research that rigorously evaluates the accuracy of the predictive models and their effects on health outcomes when used in particular ways in real clinical settings. Benefits of incorporating AI in our routine healthcare system include increased efficiency in treating and diagnosing patients by giving physicians the ability to focus on diagnoses and procedures that require more remarkable skill and judgment, increased diagnostic accuracy, and improved treatment regimens.
However, there are also several risks associated with integrating AI into medical devices used to diagnose and treat patients, such as potential liability for harm to patients in case the decision made by the algorithm is incorrect, unauthorized use of private health information, and chances for reduced physician and patient medical decision making.
Despite all these arguments, the reality is that AI will continue to grow in the medical field and alter existing workflows, paradigms, and relationships.