Artificial intelligence (AI) solutions that automatically extract information from digital histology images have shown great promise for improving pathological diagnosis. Prior to routine use, it is important to evaluate their predictive performance and obtain regulatory approval. This assessment requires appropriate test datasets. However, compiling such datasets is challenging and specific recommendations are missing. A committee of various stakeholders, including commercial AI developers, pathologists, and researchers, discussed key aspects and conducted extensive literature reviews on test datasets in pathology. Here, we summarize the results and derive general recommendations for the collection of test datasets. We address several questions: Which and how many images are needed? How to deal with low-prevalence subsets? How can potential bias be detected? How should datasets be reported? What are the regulatory requirements in different countries? The recommendations are intended to help AI developers demonstrate the utility of their products and to help regulatory agencies and end users verify reported performance measures. Further research is needed to formulate criteria for sufficiently representative test datasets so that AI solutions can operate with less user intervention and better support diagnostic workflows in the future.

The Food and Drug Administration has approved the first artificial intelligence software to be used to interpret dental x-rays, allowing dentists to better screen for oral pathologies. Pearl's Second Opinion is the first and only FDA-cleared AI radiologic detection aid for dentists at the chairside, and it can assist dentists to discover a variety of common dental diseases such as tooth decay, calculus, and root abscesses. Pearl gathered over 100 million dental x-rays from dental practices and academic institutes to create Second Opinion. The AI platform highlights anomalies in x-rays and also acts as a patient communication tool, allowing dentists to exhibit alternative models of a patient's teeth and highlight trouble regions. Pearl's announcement is a significant step forward in the field of technology-assisted dentistry.

The past decade has seen an important and, for many patients, a life-changing rise in the number of innovative new drugs reaching the market to treat diseases such as multiple sclerosis, malaria, and subtypes of certain cancers (such as melanoma or leukemia). In the United States, the Food and Drug Administration approved an average of 41 new molecular entities (including biologic license applications) each year from 2011 to 2020--almost double the number in the previous decade. Despite the immense costs of such achievements, 2 2. Asher Mullard, "New drugs cost US $2.6 billion to develop," Nature Reviews Drug Discovery, December 1, 2014. A major barrier is the daunting challenge of understanding the multifactorial nature of many diseases coupled with the vast set of variables in therapy design. Very few diseases, such as cystic fibrosis, are linked to variants in single genes. Drug development therefore tends to rely on a reductionist, hypothesis-driven approach that narrows the focus to individual cell types or pathways. Focused assays often based on partial information or informed by animal models that never perfectly reflect human disease then attempt to identify single molecules that will benefit patients.

The Internet of Medical Things (IoMT) is rapidly gaining traction as healthcare organizations seek new methods to innovate and stay current with technology in order to help patients and clinicians better monitor and track ailments. The Internet of Things (IoT) has grown rapidly in healthcare in recent years, and the pandemic has accelerated the emergence of IoMT, which several governments have widely used as a strategy to prevent the virus's spread. The industry has one of the greatest rates of IoT investment and growth. A 2020 FICCI-BCG report states that: "Health tracking devices and electronic health records combined with AI and Machine Learning (ML) technologies will not only pave the way for efficient chronic disease management but will also lead to the emergence of efficient clinical decision support systems." The EMBLEM MRI Subcutaneous Implantable Defibrillator (S-ICD) System as well as the EMPOWER Modular Pacing System (MPS), which is planned to be the first leadless pacemaker capable of providing both bradycardia pacing support and anti-tachycardia pacing, make up the mCRM System.

Could a new artificial intelligence (AI)-enabled advance have an impact in the diagnosis and treatment of brain aneurysms? The Food and Drug Administration (FDA) has granted 510(k) clearance to Aidoc's new AI platform for brain aneurysms. In addition to identifying and triaging suspected cases, the algorithm facilitates communication and workflow between radiologists, neurologists and neuroendovascular surgeons, according to the company. Researchers have estimated that approximately 6.5 million people in the United States have an unruptured brain aneurysm.1 The Brain Aneurysm Foundation has noted that most aneurysms are small, ranging from 1/8 inch to an inch, and ruptured aneurysms are reportedly misdiagnosed or there is a delay in diagnosis in 25 percent of patients who present to health-care providers.1 Elad Walach, the CEO and co-founder of Aidoc, said the new AI-enabled algorithm may help enhance timely diagnosis and care for brain aneurysms.

This article is going to be a snapshot of some things going on in Artificial Intelligence at the intersection of healthcare. Are you an iOS User and like Online surveys? Jasmine Sun asked me to share this opportunity with you guys: Participate in a Substack reader interview. They are looking for Substack users who aren't power users. I most recently covered in the A.I. intersection of Healthcare the following topics: Artificial Intelligence is Taking on Parkinson's Disease.

An artificial intelligence tool that reads chest X-rays without oversight from a radiologist got regulatory clearance in the European Union last week -- a first for a fully autonomous medical imaging AI, the company, called Oxipit, said in a statement. It's a big milestone for AI and likely to be contentious, as radiologists have spent the last few years pushing back on efforts to fully automate parts of their job. The tool, called ChestLink, scans chest X-rays and automatically sends patient reports on those that it sees as totally healthy, with no abnormalities. Any images that the tool flags as having a potential problem are sent to a radiologist for review. Most X-rays in primary care don't have any problems, so automating the process for those scans could cut down on radiologists' workloads, the Oxipit said in informational materials.

An artificial intelligence tool that reads chest X-rays without oversight from a radiologist got regulatory clearance in the European Union last week -- a first for a fully autonomous medical imaging AI, the company, called Oxipit, said in a statement. It's a big milestone for AI and likely to be contentious, as radiologists have spent the last few years pushing back on efforts to fully automate parts of their job. The tool, called ChestLink, scans chest X-rays and automatically sends patient reports on those that it sees as totally healthy, with no abnormalities. Any images that the tool flags as having a potential problem are sent to a radiologist for review. Most X-rays in primary care don't have any problems, so automating the process for those scans could cut down on radiologists' workloads, the Oxipit said in informational materials.

There is a common misconception that once AI becomes embedded in healthcare processes, it will take away human jobs. That scenario is unlikely to occur, however, because AI's true role is to aggregate and analyze reams of data. That task is difficult for humans, so AI will be a welcome partner in clinical decision-making. This stance is supported by Scott Gottlieb, former commissioner at the Food and Drug Administration (FDA) and now a board member at Pfizer and Illumina. He is also a senior fellow at the American Enterprise Institute, a public policy firm.

Algorithms are increasingly being put to work alongside radiologists and pathologists to help detect and diagnose cancers. Why it matters: AI developers say these tools can help relieve a stressed health care system and improve critical medical decision-making, but experts caution about the risk of overdiagnosis that could drive up health spending and bring the possibility of unnecessary, risky biopsies. Where it stands: Since the FDA began regulating algorithms as medical devices a few years ago, there's been a surge in computer models developed to help detect and diagnose cancer and to assist in prioritizing the workflow of radiologists. The next frontier for AI in cancer pathology is to use biological markers "to predict a clinical outcome that we don't easily predict today," says David Klimstra, the chief medical officer and co-founder of PaigeAI. Keep in mind: FDA clearance is a lower bar than FDA approval when it comes to medical devices, including AI. Yes, but: There is a risk of false positives and unnecessary treatment.