Artificial intelligence could help improve mental health care, making it more effective, along with addressing personnel issues that are poised to impact the mental health field over the course of the next decade. AI can potentially derive complex patterns from data, patterns which even trained psychiatrists and clinicians have difficulty seeing. Furthermore, as reported by Time, AI could help make up for critical shortages in mental health care workers, giving patients support outside of the time they spend with a clinician.
Over the course of the next five years, it is predicted that the US mental health system may come to lack around 15600 psychiatrists, according to a study published by the National Council for Behavioral Health. This means that the time and resources of clinicians and other mental health professionals will be spread thin. Even now, clinicians typically don’t spend very much time with their patients, occasionally seeing a patient just once every few months.
Artificial Intelligence has recently made huge contributions to the medical field, improving accuracy of diagnosis, using computer vision to find obscure patterns in medical images, and engineering better treatment plans for patients. One of the ways mental healthcare differs from physical health care is that it requires a high level of perception and emotional intelligence in order to diagnose and treat patients, but AI could still have a beneficial effect on the field. The power of machine learning to analyze data and extract patterns, including those so confusing and subtle that humans find them hard to notice/interpret, can help mental health professionals treat and support their patients.
Advanced data analysis techniques can improve the diagnosis of certain mental health conditions like bipolar disease, and the quicker the diagnosis the faster patients can be put on the proper treatment course. Beyond this, artificial intelligence can help clinicians in other ways, such as letting doctors interact with their patients remotely or by automatically collecting and analyzing data that can be used to update treatment plans.
Currently, there are some applications that leverage artificial intelligence to support those with mental health conditions. Woebot, for instance, is a chatbot that employs principles derived from cognitive behavioral therapy to assist people in tracking their moods and managing thought patterns. The coming years could see much more sophisticated applications of artificial intelligence in the mental health field. As. Dr. Henry Nasrallah, a psychiatrist at the University of Cincinnati Medical Center explained to Time, there are methods that can be used to infer a patient’s mental health state, such as lack of speech affectation often correlating with depression, or with disjointed word use being linked to schizophrenia.
Clinicians often use speech patterns like this to diagnose patients, and AI algorithms can potentially pick up on patterns so subtle that people won’t discover them. Recently Peter Foltz, a research professor at the University of Colorado Boulder, and his collleagues created an app that has patients undergo various verbal exercises, collecting data about their tone and affections as they answer questions regarding their emotional state and relate stories. This data is then analyzed by an AI system that compares the clips against sound bites from a larger patient population in order to discover potential mental health issues. When tested on a population of 225 individuals across two different locations, the app performed at least as well as clinicians at detecting symptoms of mental health disorders or distress. Similar pattern recognition can also be done with written language, analyzing word choice and order of word use.
There are a few notable roadblocks to creating AI-based mental health diagnosis tools. One of the biggest issues is that clinicians and psychiatrists themselves often disagree about what criteria is needed to make a diagnosis, with illnesses like depression based on a variety of scales and criteria. Other problems like the questionable reliability of patient-reported data could also hamper efforts to design diagnostic AI tools. Even AI and mental health researchers themselves emphasize that their tools aren’t meant to replace human psychiatrists and acknowledge their limitations. However, as data collection gets better and models become more sophisticated, the reliability of mental health diagnostic AIs may increase. Finally, by automating many of the time-consuming processes that clinicians must deal with, AIs can allow mental healthcare providers to spend more time with patients, a worthy goal all on its own.
Paper Examines How To Reduce Risk Of Using AI in Medicine
Artificial intelligence programs are capable of improving healthcare in a variety of different ways. For instance, AI applications can use computer vision to help doctors diagnose conditions from X-rays and FMRIs. Machine learning algorithms can also be used to help reduce false-positive rates by extracting subtle patterns from data that humans may not be able to find in medical data. However, with the possibilities comes new challenges, and recently a new article was published in Science that examined possible risks and regulatory strategies for medical machine learning techniques in an effort to minimize any possible negative side effects of employing AI in a medical context.
Expanding Applications For AI In Healthcare
AI is seeing its applications in the medical field expand rapidly. Recent developments in the field of healthcare, driven by AI, include the creation of a new pharmaceutical company that aims to use AI to create new drugs, the creation of AI-drive remote health sensors, and computer vision apps that analyze CT scans and X-rays.
To be more precise, Genesis Therapeutics is a startup that is aiming to use AI to speed up the process of drug-discovery, hoping to create drugs that can reduce the severity of debilitating diseases. Genesis Therapeutics is just one of almost 170 different firms using AI to research new drug formulations. Meanwhile, in terms of health monitoring devices, iRhythm and French AI startup Cardiologs are making use of AI algorithms to analyze EEG data and monitor the health of those who have heart conditions are at risk of complications. The software designed by the companies can detect heart murmurs, a condition caused by turbulent blood flow.
Finally, a recent study investigating how computer vision can be applied to medical images found that computer vision systems perform at least as well or better than expert radiologists when examining CT scans to find small hemorrhages. The algorithms used in the study were able to render predictions after examining CT scans for just one second. The computer vision systems were also able to localize the hemorrhage within the brain.
So while the potential benefits of using AI in healthcare are clear, what is less clear is what new challenges and risks will arise as a side-effect of employing AI within the healthcare field.
Regulating An Expanding Field
As TechXplore reported, in order to assess potential drawbacks of using AI in healthcare, a group of researches recently published a paper in Science, aiming to derive answers to anticipate potential problems with AI and explore potential solutions to these problems. Problems that may arise from using AI in the healthcare field include the inappropriate recommendation of treatments resulting in injury, privacy concerns, and algorithmic bias/inequality.
The FDA has only approved medical AI that uses “locked algorithms”, algorithms that reliably produce the same result every time they are run. However, much of AI’s potential lies in its ability to learn and respond to new types of inputs. In order to enable “adaptive algorithms” to see more use and get approval from the FDA, the authors of the paper took an in-depth look at how the risks related to updating algorithms can be mitigated.
The authors advocate that machine learning engineers and researchers should focus on continuous monitoring of models over the lifetime of their deployment. Among the suggested tools to monitor AI systems was AI itself, which could help give automated reports on how an AI is behaving. It’s also possible that multiple AI devices could monitor each other.
“To manage the risks, regulators should focus particularly on continuous monitoring and risk assessment, and less on planning for future algorithm changes,” said the authors of the paper.
The authors of the paper also recommend that regulators focus on developing new methods of identifying, monitoring, assessing, and managing risks. The paper applies many of the techniques that the FDA has used to regulate other forms of medical tech.
As the paper’s authors explained:
“Our goal is to emphasise the risks that can arise from unanticipated changes in how medical AI/ML systems react or adapt to their environments. Subtle, often unrecognised parametric updates or new types of data can cause large and costly mistakes.”
Benjamin Sexson, CEO Monogram Orthopaedics – Interview Series
Benjamin Sexson is the CEO of Monogram Orthopedics. Prior to joining Monogram, Mr. Sexson served as the Director of Business Development at Pro-Dex, an OEM manufacturer of Orthopedic Robotic End-Effectors. In his tenure at Pro-Dex, Mr. Sexson was responsible for the development, management, and launch of a proprietary product solution, helping to negate a distribution agreement with a major strategic partner.
Could you explain the mission statement of Monogram Orthopaedics?
Our Mission is to make orthopaedics personal. The current standard of care is highly impersonal. In crude terms, patients are permanently and irreversibly amputating arthritic bone to have it replaced with an off-the-shelf generic implant that, in non-clinical terms, gets “hammered” into place. We quite literally are replacing joints with implants that are guaranteed not to fit perfectly. We are working to mitigate the risks of arthroplasty with technology that drives personalization.
What type of scans are necessary in order to fully design the 3D molding for the knee or hip replacement?
We are utilizing computed tomography scans (also CT scans). Optimally the CT examination is performed at 140 kV and 300 mA with slice thicknesses of 0.625mm in regions of interest and thicknesses of 3-5mm in areas where we aren’t reconstructing anatomy with an implant.
How is machine learning used to parse the data from these scans?
Bone is a composite. It consists of compact bone at the periphery, spongy bone, and bone marrow. Our implants are designed to maximize contact with the inner cortical wall (inner surface of the compacted bone at the periphery) to improve initial stability.
The primary purpose of the machine learning algorithms is to segment bone (inner and outer cortex) from the CT scans as well as to identify the critical anatomical landmarks that inform the implant design algorithms.
Where does the 3D printing take place?
The FDA mandates highly rigorous quality standards. Initially, we will be using an ISO 13485 contract manufacturer. As we scale, we will vertically integrate.
Monogram will leverage the kinematic redundancy of robotics to allow for intraoperative obstacle avoidance and real-time tissue tracking. Will this decrease the risks of surgery complications such as blood clots or infection?
Our innovation starts with the implants – we are driving improved fixation in a smaller form factor. Infections with implants can be especially insidious and a leading cause of revision. Generally, once a thin, slimy film of bacteria commonly referred to as biofilm adheres to the implant surface, the infection is irreversible, and the implant removal is required. By reducing the size of our implants, we reserve bone stock for future revision. Furthermore, we hypothesize that our increased cortical contact will facilitate more even and proximal loading of the femur, mitigating the risk of stress shielding.
Notably, Monogram has commissioned research in collaboration with a pharmaceutical company to explore a proprietary coating with very high adherence that may reduce the risks of infection without being cytotoxic.
Will this speed the healing time for patients?
It is unlikely to reduce the healing time for patients, but it is likely to mitigate the risks of loosening, fracture, and poor placement. Improved initial fixation can lessen pain from implant migration. We also believe we will preserve more bone stock in the adverse event a revision is needed. Our implants rely on natural, long term fixation, which is especially favorable for younger active patients.
You recently chose to fundraise a Series A via the SeedInvest crowdfunding platform. What inspired you to crowdfund versus traditional VC funding?
Monogram needs a significant amount of capital to navigate the FDA approval process and commercialize our technology. The problem with traditional VC financing is that a financing round for large deals generally needs to be fully syndicated before a company has access to capital. With SeedInvest, it’s more like a shelf-offering. Once we hit our escrow target of $2.75M, we have access to those funds immediately, allowing us to hire talent and continue our development efforts as we fundraise. Furthermore, we are laying the groundwork to go public down the road, so this is highly strategic for the long term.
Monogram is in the process of onboarding a new VP of Engineering and Director of Implant Engineering. The next steps are finalizing discussions we are having with a strategic partner for licensing the generic components of our design (for example, the tibial locking mechanism) and completing the testing we are running on our knee with the University of Nebraska. Our next major milestone will be our FDA submission and from there an FDA approval.
Is there anything else that you would like to share regarding Monogram Orthopaedics?
We are commercializing highly complex technology and there is a bit of a learning curve to the industry and our technology. To accommodate people who are interested in learning more we regularly host webinars with live Q&A’s as a management team. More information on future webinars is available by clicking here.
London-Based Startup LabGenius Raises $10M
The London-based startup LabGenius announced that they raised over $10 million in Series A Funding. They are a drug discovery company that utilizes artificial intelligence (AI), robotic automation, and synthetic biology. Their main focus is to find novel protein therapeutics.
According to Dr James Field, CEO and Founder of LabGenius, “Protein therapeutics have an unparalleled potential to both treat disease and alleviate human suffering. By transforming how these drugs are discovered, we have a shot at improving the lives of countless people. Being able to robustly engineer novel therapeutic proteins has immense commercial and societal value. The discovery of protein therapeutics has historically been highly artisanal, relying heavily on humans for both experimental design and execution. This dependence has proved limiting because, as a species, we’re cognitively incapable of fully grasping the complexity of biological systems.”
The Series A investment round is led by Lux Capital and Obvious Ventures. Other participants included Felicis Ventures, Inovia Capital, Air Street Capital, and other existing investors. CEO and founder of Recursion Pharmaceuticals, Chris Gibson, along with Inovia Capital General Partner Patrick Pichette, are also investing. Pichette is the former CFO of Google.
According to the company, they will use the capital to “scale their team, expand the scope of its discovery platform, and initiate an internal asset development program.” One of their main goals is to evolve novel antibody fragments. These could be used to treat certain conditions that can’t rely on conventional antibody formats.
Lux Capital and Obvious Ventures
Zavain Dar, Partner at Lux Capital, along with Nan Li, Managing Director at Obvious Ventures, have been involved in the life science startup environment for some time. Their investment strategy dates back nine years, including a 2013 investment into Zymergen, a molecule discovery and manufacture company based out of California. In 2016, they were involved in Recursion Pharmaceuticals, who went on to a series C raise of $156 million in July.
Their strategy follows a path, starting at industrial biotech technology with Zymergen and followed by root-cause biology discovery with Recursion Pharmaceuticals. It is closed out by creating composition of matter and IP with LabGenius.
Dar explained his reasoning behind choosing LabGenius over other startups.
“We scoured the globe, and didn’t want to be constrained by what happened to be in our backyard,” he says. “They are leading the pack…and now with backing and pharma partnerships, should be in a good position.”
Humans No Longer Sole Agents of Innovation
When speaking to TechCrunch, Dr James Field said, “My central thesis, the thing that’s really the driving force behind the company, is the conviction that we’re entering an age in which humans will no longer be the sole agents of innovation. Instead, new knowledge, technologies and sophisticated real-world products will be invented by smart robotic platforms called empirical computation engines. An empirical computation engine is an artificial system capable of recursively and intelligently searching a solution space.”
The company has created a discovery platform called EVA, and it integrates multiple new technologies coming from different fields including artificial intelligence. After discovery and characterisation, LabGenius then sends their proprietary molecules to clinics.
Field explains the company’s EVA technology as a “machine learning-driven, robotic platform”,” that is capable of “designing, conducting and critically learning from its own experiments.”
“For decades, scientists, engineers and technologists have dreamt of building ‘robot scientists’ capable of autonomously discovering new knowledge, technologies and sophisticated real-world products,” says Field.
“For protein engineers, that dream has now entered the realm of possibility. The rapid pace of technological development across the fields of synthetic biology, robotic automation and ML has given us access to all the essential ingredients required to create a smart robotic platform capable of intelligently discovering novel therapeutic proteins.”