Healthcare and artificial intelligence are two fields that continue to be targets for a huge investment and have been topping industrial investment for 4 years running. Due to the continuing COVID-19 pandemic, focus on telehealth, personalized care, and bias management are receiving intense focus. Here are some predictions we have for the upcoming year. Recognizing that algorithmic bias has life and death consequences in healthcare, AI development is leading to very specific AI/ML model deployment.
COVID-19 is changing the medical landscape. Omicron is the primary variant of COVID-19 currently with 300,000 new cases a day in the United States, a new high. The ongoing pandemic forces us all to consider how we are preparing for the next variant, and then, the next pandemic.
-
Telemedicine is sticking around as COVD-19 Continues… 2021 was the year that continued to prove the importance of telemedicine.The United States is averaging more than 300,000 new cases a day for the first time in the pandemic due to the new Omicron variant though hospitalizations are growing at a much slower rate than cases [1] . Telehealth has emerged to the COVID-19, improving access to care and pandemic due to transition toward digital medicine. We believe 2022 will cement these gains now and could help make healthcare more equitable once the pandemic has ended. [2] .
-
Models are utilizing AI to assess COVID-19 images. Rapid identification of COVID-19 and its severity in individual patients is expected to enable better control of the disease individually and at-large. There has been remarkable interest by the scientific community in using imaging biomarkers to improve detection and management of COVID-19. Exploratory tools such as AI-based models may help explain the complex biological mechanisms and provide better understanding of the underlying pathophysiological processes. The present review focuses on AI-based COVID-19 studies as it applies to chest x-ray (CXR) and computed tomography (CT) imaging modalities, and the associated challenges [3, p. 19] .
-
Personalized healthcare is becoming the darling of the industry . This is not limited to generic diagnoses according to sex and race. Understanding individual risk at the level of DNA and socioeconomic factors are leading to a renewed interest in clinical and artificial intelligence research [4]–[6] .
-
Specific use cases or bust. We’re not at the point where we have robot doctors yet. Startups, research institutions and larger organizations alike are starting small with very specific use cases. For example, it can take a chest CT and find pulmonary nodules or pneumothorax or pulmonary emboli.
-
AI Predicts Cancer: Models designed to detect cancer at earlier stages are becoming more prevalent. MIT researcher Regina Barzilay and her team created the mammogram detection program Mirai. It could transform how mammograms are used, open up a whole new world of testing and prevention, allow patients to avoid aggressive treatments and even save the lives of countless people who get breast cancer. Mirai would spit out risk scores for patients’ next five years, giving them a chance to make health-care choices that earlier generations could only dream of [7] .
-
Venture Capital will continue huge healthcare innovation investments in 2022 . 2020 and 2021 have been strong years for healthcare startups. Healthcare startup investment 2020 exceeded 2019 with 52B. 2020 4th straight record year of healthcare VC fundraising and 2021 was not far behind. HealthTech investments set a new record this year in each of its subsections with biopharma taking the lead with 24.6B in venture capital investment; $1B total raised in medical devices, $5.9B in alternative care, $3.7B in provider operations, $2B in clinical trial enablement. IPOs in each sector set or tied annual highs while Dx/tools built have out performed their IPO by 100%.With artificial intelligence applications such as computer vision and automation assisting startups in achieving efficiency in simpler ML applications, 2022 [8]
-
Dataset assembly will continue to be the backbone of all medical research and innovation . The biggest hurdle that currently exists for clinical and commercial implementations of AI models in healthcare is the lack of easily accessible diverse datasets to train models with. Gradient Health is committed to developing off the self clinically-ready datasets to spur on the healthcare revolution.
Gradient Health, Inc is a healthcare startup with product offerings in dataset assembly and radiologist annotation tools and eventually, clinical trials. Gradient Health currently produces labeled datasets of medical imagery, taking raw DICOM data and refining it with physician-created labels to assist in machine learning.
Citations
