Blog Tag: artificial intelligence

AI and Cancer Diagnostics

Rather than simply reading radiological scans, Ibex Medical Analytics is employing its artificial intelligence (AI)-powered cancer detection solutions to assist pathologists in the lab. Ibex’s Galen Platform is a clinical-grade, multi-tissue platform that helps pathologists detect a variety of cancers and grade their malignancies.  Ibex states that the platform can also “detect more than a hundred other clinically relevant diagnostic features in multiple tissue types.” Ibex leverages data extracted from millions of pathology slides along with AI and machine learning to perform its platform’s diagnostic functions.

Ibex reports that its breast cancer diagnostic tool, Galen Breast, delivered clinical grade accuracy in diagnosing multiple breast cancer types in a recent multi-site clinical study conducted at Institut Curie in France and Maccabi Healthcare Services in Israel. The study compared the performance of pathologists using Ibex AI against pathologists using traditional microscopes to diagnose multiple types of breast cancer. The study results showed the Galen Breast tool could detect various forms of breast cancer at a level of accuracy on par with trained pathologists. According to an Ibex press release, Galen Breast technology has shown the ability to detect both invasive and in-situ cancers in breast biopsies, distinguish between multiple sub-types of cancer (e.g., lobular versus ductal carcinoma), grade cancer severity in situ, and identify rare tumors. The Galen Platform additionally offers AI solutions for the detection of prostate and gastric cancers, which Ibex states also deliver clinical grade accuracy.

More recently, the company announced the launch of its latest Galen Platform version, Galen 3.0.  Ibex reports that Galen 3.0 provides expanded detection capabilities and a broader set of features to support pathologists in diagnosing breast, prostate, and gastric cancers.

 

FDA Expresses Priorities for Clinical Trial Efficiency, Artificial Intelligence

The FDA has announced new goals to help modernize its procedures and respond to new technologies.  In a blog post by FDA Commissioner Scott Gottlieb, M.D., the agency expressed new priorities to help modernize clinical trials for medical devices and develop standards for new technologies like artificial intelligence.

According to Gottlieb, clinical trials “are becoming more costly and complex to administer” while “new technologies and sources of data and analysis make better approaches possible.”  In order to take advantage of these better approaches, Gottlieb pointed to the FDA’s Breakthrough Devices Draft Guidance, which proposes streamlined procedures to develop flexible clinical trial designs for important medical devices.  This will allow the FDA to “evaluate . . . innovative devices more efficiently.”  Six breakthrough devices have already been cleared using this program.

Additionally, Gottlieb discussed the FDA’s new goal of enabling the use of “real-world evidence” to support decisions to approve devices.  According to Gottlieb, “[r]eal world evidence can help answer questions that are relevant to broader patient populations or treatment settings where information may not be captured through traditional clinical trials.”  The FDA is helping to design several proof-of-concept trials that utilize real-world evidence.

Finally, Gottlieb discussed the FDA’s role in dealing with new and emerging technologies.  In particular, Gottlieb discussed artificial intelligence, which “holds enormous promise for the future of medicine.”  Medical artificial intelligence models are currently in development and the FDA recently approved an AI algorithm for detection and treatment of distal radius fractures.  According to Gottlieb, the FDA is exploring ways to handle and evaluate the kinds of data that are relevant to AI performance and safety, hoping to “enable a transparent benchmarking system for AI algorithm’s performance.”

Gottlieb concludes that the FDA has “undertaken a comprehensive effort to make sure that our organization and policies are as modern as the technologies we’re being asked to evaluate, and that we’re able to efficiently advance safe, effective new innovations.”

Veterans Association and IBM Watson Extends Oncology Partnership

On July 19th, 2018, the United States Department of Veterans Affairs (VA) and IBM Watson Health announced an extension of their ongoing partnership to interpret cancer data in patients.

According to the press release, the partnership was originally announced two years ago with the goal of providing precision care for veterans. Oncologists and pathologists receive tumor samples from patients nationwide. The DNA is sequenced from these samples. Watson then interprets the genomic data and identifies relevant mutations. Once the mutations are identified, Watson suggests potential targeted treatment options. Watson accomplishes this by cross-referencing the findings against medical literature on potential and FDA approved therapies.

The partnership builds upon other genomic efforts by the VA. One such program is the VA’s Million Veteran Program.   Starting in 2011, this program aimed to create a massive medical and genomic database by collecting blood samples from 1 million volunteers. The VA noted that by taking baseline and subsequent readings based on military experiences, health and lifestyles, the information contained in the database could hold the key to preventing and treating diseases.

According to press releases, the VA currently treats 3.5% of the United State’s cancer patients. This makes the VA the largest health care provider treating cancer in the United States.  According to the National Cancer Institute, in 2018, an estimated 1,735,350 new cases of cancer will be diagnosed in the United States.

In its press release, Dr. Kyu Rhee, chief health officer for IBM Watson Health. “It is incredibly challenging to read, understand and stay up-to-date with the breadth and depth of the medical literature, and link them to relevant mutations for personalized cancer treatments. This is where AI can play an important role in helping to scale precision oncology, as demonstrated in our work with VA, the largest integrated health system in the U.S.”

 

FDA grants De Novo Market Clearance of Artificial Intelligence Software

According to a U.S. Food and Drug Administration press release, Viz. AI Contact application was granted De Novo premarket review to Viz.AI’s LVO Stroke Platform. According to PR Newswire, Viz.AI’s LVO Stroke Platform is the “first artificial intelligence triage software” and its approval begins “a new era of intelligent stroke care begins as regulatory approval.” The Viz.AI LVO Stroke Platform, according to the U.S. Food and Drug Administration press release, is a clinical support software designed to analyze Computerized Tomography (CT) scans, identify suspected large vessel blockage, and send a notification to specialist of a potential stroke in patients sooner.

According to the Centers for Disease Control and Prevention, strokes are the fifth leading cause of death in America. A stroke occurs when the blood vessels in the brain are damaged, compromising the necessary blood flow to the brain. There are many types of strokes and can often lead to brain damage, long term disability, and death. A CT scan can show the location and extent of the damage to the brain to diagnose the stroke as well as the type of stroke that has occurred.

Viz.AI is a healthcare company based in San Francisco and Tel Aviv, dedicated to “expand Direct-to Intervention care” which “advances information about treatable patients straight to the interventionalist.” Neurosurgeon and CEO of Viz.Al, Dr. Chris Mansi stated in a press release:

“The Viz.ai LVO Stroke Platform is the first example of applied artificial intelligence software that seeks to augment the diagnostic and treatment pathway of critically unwell stroke patients.”

According to the FDA press release, Viz. AI Contact application was granted De Novo premarket review, which is a “regulatory pathway for new types of medical devices that are low to moderate risk and have no legally marketed predicate device to base a determination of substantial equivalence.” This is a new regulatory classification, “which means that subsequent computer-aided triage software devices with the same medical imaging intended use may go through the FDA’s premarket notification (510 (k)) process, whereby devices can obtain marketing authorization by demonstrating substantial equivalence to a predicate device.”

The Viz.AI Contact application is one example of what the FDA calls “clinical decision support software (CDS). CDS includes technology that aids in diagnosing and identifying treatment plans. CDS includes “technology has the potential to enable providers and patients to fully leverage digital tools to improve decision making.” The FDA is currently creating a regulatory framework for CDS to provide guidance and encourage developers in this field.

According to Robert Ochs, acting deputy director for radiological health, Office of In Vitro Diagnostics and Radiological Health in the FDA’s Center for Devices and Radiological Health, “(This) software device could benefit patients by notifying a specialist earlier thereby decreasing the time to treatment. Faster treatment may lessen the extent or progression of a stroke.”

MedyMatch Intracranial Hemorrhage Detection Software Receives Expedited FDA Review

MedyMatch Intracranial Hemorrhage Detection Software Receives Expedited FDA Review

Tel Aviv-based MedyMatch Technology recently announced it has received Expedited Access Pathway (EAP) designation from the FDA for its intracranial hemorrahage (ICH) detection software medical device.

According to Gene Saragnese, Chairman & CEO of MedyMatch, the platform is a “first-in-class hemorrhage detection tool.”  The MedyMatch device utilizes artificial intelligence and deep learning technologies to analyze non-contrast head CT images for signs of ICH.  Further implementations of the MedyMatch deep vision platform include diagnosis and monitoring of acute and chronic diseases based on concurrent analysis of imaging data and other patient data.  Vice President of Clinical, Regulatory, and Quality Affairs, Dr. Joshua Schulman, said:

This designation is a recognition of both the need for new assessment tools for intracranial hemorrhage and an affirmation of MedyMatch’s technical approach to assisting clinicians to need to make time-sensitive yet accurate decisions in emergency settings.

The EAP Program, intended to speed approval of certain medical devices, generally includes priority review, more interactive review, and senior management involvement.  EAP designation can be awarded for devices that address unmet needs for treatment or diagnosis of life-threatening or irreversibly debilitating conditions.    It is said to be expected that EAP designated devices will be transitioned to the new Breakthrough Devices program established under the 21st Century Cures Act of 2016.

Digital Currencies and Blockchain in the Medical Arena

Recently, digital currencies, such as bitcoin, have greatly increased in popularity. Some of this popularity may be attributed to digital currencies’ many purported advantages over traditional currencies, such as that blockchain technology allows for a distributed and cryptographically secure ledger without the use of traditional banking institutions. Newer and more advanced digital currencies have recently been introduced with the added advantage of smart contracts, which are said to be self-executing contractual clauses that may be programmed into a digital currency transaction. As such, many new digital currencies have been appearing with individuals investing in Initial Coin Offerings (ICOs), which are somewhat akin to the Initial Public Offerings (IPOs) of a traditional corporation.

Even more recently, a few companies have begun to make use of digital currencies and blockchain technology in the medical arena. Many have found blockchain technology uniquely suited to secure patient records, and have found that the smart contracts of digital currencies may allow individuals greater control of their medical data. Below is a summary of a few fields of medicine and companies within those fields in which digital currencies and blockchain are already being developed.

Medical Records and Health Data

According to The Merkle, Bowhead Health is the first medical device company using their AHT digital currency tokens with smart contracts to create a new medical data market. The company plans to allow individuals with Bowhead’s digital currency to control the dissemination of their medical data, and also to compensate those individuals if and when they choose to share with research institutions. Bowhead’s AHT tokens are said to allow 70% of research fees to be distributed to users with the other 30% going to token holders.

According to Blockchain News, Medicalchain is a UK-based company using blockchain technology to allow patients to securely store and send their medical records to their healthcare professionals. Medicalchain is said to allow patients to have a centralized medical record accessible from anywhere in the world, and allow individuals the ability to control medical institutions’ access to their records.

The Medical Society of Delaware has partnered with the company Medscient, and they are using blockchain technology to create a proof-of-concept platform to allow insurers and medical care providers to access patient records, according to The Cointelegraph. The article further states that this partnership was made possible when the state of Delaware became the first state to pass a law allowing the use of blockchain technology in business for stock trading and record-keeping.

Medical Licenses

The Illinois Blockchain Initiative has partnered with Hashed Health to create a pilot program to streamline the process of issuing and tracking medical licenses, according to The Cointelegraph. The goal of this partnership is said to give patients and healthcare providers a transparent license registry system that uses smart contracts to automatically update information.

Medicine and Artificial Intelligence (AI)

According to news sources, Doc.ai is a collaboration between developers from the universities of Stanford and Cambridge, and is said to be creating a platform built on blockchain technology and using AI to create a resource to answer patient’s specific questions regarding their personal health records and their physician’s analysis.

Medical Device Compared to ‘Star Trek Tricorder’ Presented at AACC

Medical Device Compared to ‘Star Trek Tricorder’ Presented at AACC

Basil Leaf Technologies recently presented their DxtER device at the 69th AACC Annual Scientific Meeting & Clinical Lab Expo in San Diego.  The DxtER device has been compared to the Star Trek medical Tricorder, winning first place in the Qualcomm Tricorder Xprize competition, a global contest inspired by the popular science fiction series.

According to Basil Leaf Technologies, the DxtER device weighs less than five pounds and is designed to enable consumers to monitor five real-time health vital signs and diagnose 34 diseases using artificial intelligence.  Basil Leaf Technologies reports that the DxtER device is currently undergoing clinical trials for FDA approval.

According to AACC CEO, Janet B. Kreizman, “DxtER is the first consumer-friendly mobile health device to combine vital sign monitoring with an extensive diagnostic testing menu, and it could lead to a huge leap forward in patient care.”  While the DxtER device may not be available in the immediate future, Dr. Gene Friedman, assistant professor of biomedical engineering at Johns Hopkins University School of Medicine, estimates that “in the next 10 to 20 years it’s going to be a big revolution in personal healthcare.”

FDA to Create New Digital Health Unit

The FDA has announced that it will create a new unit responsible for overseeing digital health products.  According to Bakul Patel, the Associate Center Director for Digital Health at FDA, the agency is working to define “what clinical validation looks like for software,” and will consider creating new premarket approval or clearance procedures for digital health products.

As Patel stated in an interview with Wired, the new unit will aim to keep up with emerging technology that uses artificial intelligence and machine learning to diagnose and treat patients.  Patel noted that this kind of technology creates new and difficult challenges for regulators, who must figure out how to ensure that patients remain safe while keeping up with the increasing complexity and tighter release schedules that these new kinds of software bring.

“We’ve been trying to translate the current regulation paradigm for digital . . . [b]ut what we have today and what we’re going to have tomorrow are not really translatable. We need to take the blinders off, start with a clean sheet of paper.” –Bakul Patel, Associate Center Director for Digital Health, FDA

Wired reports that Patel is working to develop new models for testing and approving digital health products.  Patel told Wired that “the idea is to get safe products to market faster, by having people compete on excellence rather than compliance.”  One way the FDA might do this is by providing faster review with lower scrutiny for manufacturers with a long track record of safety and compliance.

According to FierceHealthcare that the new digital health unit will be part of the FDA’s Center for Devices and Radiological Health.  While the new unit won’t be officially open until October, hiring of key staff is underway.  Wired notes that the unit will be funded by the Medical Device User Fee program.