InstaDeep, a startup from Tatooine, was acquired for $682 million by BioNTech -- er, make that Tataouine, the city in Tunisia that Tatooine, Luke Skywalker's home planet in the fictional universe of Star Wars, was named after.

"When the Covid-19 pandemic ground the world to a halt, InstaDeep trained a large language model to accurately predict new, dangerous variants before they spread."

"We took transformers already pre-trained on all proteins that existed and then altered the language of SARS-CoV-2. We tried to see if we could predict to some level of accuracy whether or not something could be high-risk. It turned out that it worked. All the variants we identified as potentially dangerous were later confirmed as concerning by the World Health Organization."

How InstaDeep became Africa’s biggest AI startup success

#solidstatelife #ai #genai #llms #medicalai

A company called Piramidal is making a "foundation model" for electroencephalography (EEG).

There doesn't seem to be much information about how the model works, other than that it's based on a similar architecture as large language models (LLMs).

Piramidal’s foundation model for brain waves could supercharge EEGs

#solidstatelife #ai #medicalai

"SiPhox Health BiomarkerAI: Transform your PDF blood test results into simple, actionable insights".

Last time I had a blood test, it seems the explanations that came with the results were pretty decent. But next time I have a blood test, I'll give this a shot. If you try it, let me know how it goes. I'm interested if you get anything better than just asking ChatGPT to explain things and recommend "actionable" actions.

SiPhox Health BiomarkerAI: Transform your PDF blood test results into simple, actionable insights

#solidstatelife #ai #medicalai

Plant Aid is an AI designed specifically for diagnosing sick plants. I don't have plants to diagnose so if you use this, let me know how it goes. They say you upload 2-3 photos and write a text description of the symptoms.

Plant Aid

#solidstatelife #ai #computervision #medicalai

Agent Hospital is a simulacrum of hospital with evolvable medical agents alrighty then. And an excuse to use the word "simulacrum".

"Once arrived the Agent Hospital, the patient's journey begins at the triage station. Patients arrive and describe their symptoms to the nursing agents. The instructions guide the nursing staff in their decision-making, enabling them to direct patients to the appropriate specialist departments where medical professional agents are available to conduct further diagnostics."

"After the initial assessment, patients follow the advice from the triage station and proceed to register at the registration counter. They then wait in the designated waiting area for their consultation turn with the specialists from the respective departments."

"When it is their turn for consultation, patients engage in a preliminary dialogue with the physician agents to describe their symptoms and the duration since onset. The physician then determines which medical examination is needed to investigate the cause and assist with diagnosis and treatment. In the current version, only one type of medical examination will be conducted for each patient based on the decisions made by doctor agents."

"After receiving the prescribed list of medical examinations, patients proceed to the relevant department to undergo the tests. The resulting medical data which are pre-generated by LLM are subsequently presented to the patient and the doctor. This process designed to mimic real-time diagnostic feedback, aligns with the presentation of symptoms."

"Subsequent to the medical examination, patients are guided to the respective department where physician agents undertake the diagnostic process. Patients disclose their symptoms and share the results of the medical examination with the physician agents, who then undergo diagnostic processes based on a predefined disease set. The diagnostic result is promptly communicated back to the patient, showcasing the model's capacity to integrate complex medical data and its advanced diagnostic ability."

"The medical agent is presented with the patient's symptoms, results from medical examinations and the diagnosis of the disease they made. In addition, three distinct treatment plans tailored to mild, moderate, and severe conditions are also provided. The doctor is then tasked with selecting the appropriate plan from the mild, moderate, or severe options, according to the patient's specific needs. If any medicine is prescribed, patients proceed to the dispensary to collect it."

"At the end of the diagnostic and treatment process, the patient provides feedback or updates on their health condition for follow-up actions. To mimic the dynamic progression of diseases accurately, the LLM-enhanced simulation involves a few key steps: doctors devise treatment plans based on the patient's detailed health information and test results, and then these details -- specifically the patient's symptoms, the prescribed treatment plan, and the diagnosed disease are incorporated into a template for simulation."

Ok, as you can see, quite an elaborate simulation. But how do the medical agents actually learn? The whole point of doing all this is to get medical agents that actually learn. Here's what they say (big chunk of quotes to follow):

"Doctor agents continuously learn and accumulate experience during the treatment process in Agent Hospital, thereby enhancing their medical capabilities similar to human doctors. We assume that doctor agents are constantly repeating this process during all working hours."

"Apart from improving their skills through clinical practice, doctor agents also proactively accumulate knowledge by reading medical documents outside of work hours. This process primarily involves strategies to avoid parametric knowledge learning for agents."

"To facilitate the evolution of LLM-powered medical agents, we propose MedAgent-Zero strategy MedAgent-Zero is a parameter-free strategy, and no manually labeled data is applied as AlphaGo-Zero."

"There are two important modules in this strategy, namely the Medical Record Library and the Experience Base. Successful cases, which are to be used as references for future medical interventions, are compiled and stored in the medical record library. For cases where treatment fails, doctors are tasked to reflect and analyze the reasons for diagnostic inaccuracies and distill a guiding principle to be used as a cautionary reminder for subsequent treatment processes."

"In the process of administering treatment, it is highly beneficial for doctors to consult and reference previously validated medical records. These medical records contain abundant knowledge and demonstrate the rationale behind accurate and adequate responses to diverse medical conditions. Therefore, we propose to build a medical record library for doctor agents to sharpen their medical abilities, including historical medical records from hospital practices and exemplar cases from medical documents."

"Learning from diagnostic errors is also crucial for the growth of doctors. We believe that LLM-powered medical professional agents can engage in self-reflection from these errors, distilling relevant principles (experience) to ensure correct diagnoses when encountering similar issues in future cases."

"If the answer is wrong, the agent will reflect the initial problem, generated answer, and golden answer to summarize reusable principles. All principles generated are subject to a validation process. Upon generation, the principle is integrated into the original question which was initially answered incorrectly, allowing medical professional agents to re-diagnose. Only if the diagnosis is correct will the principle be added to the experience base."

"To eliminate the influence of noise and maximize the utilization of the experience base, we incorporate additional judgment when utilizing experience. This judgment involves evaluating whether the top-K experience retrieved based on semantic similarity are helpful for the treating process. Helpful experience will be incorporated into the prompt, while unhelpful experience will be excluded."

Ok, so, kind of analogous to how our chatbots are originally pretrained (by self-supervised training) transformers that get further training from a reinforcement learning system called RLHF (reinforcement learning through human feedback), here we also have a LLM-based system where reinforcement learning is employed (albeit in a different way) to further train the LLMs.

I have mixed feelings about this. There's part of me that says this is a silly exercise, unlikely to produce anything reliable enough to be useful, and another part of me that says, yeah, but this could be the beginning of how all hospitals are run 20 or 30 years in the future.

Agent Hospital: A simulacrum of hospital with evolvable medical agents

#solidstatelife #ai #genai #llms #medicalai #reinforcementlearning #rl

"Xaira, an AI drug discovery startup, launches with a massive $1B, says it's 'ready' to start developing drugs."

$1 billion, holy moly, that's a lot.

"The advances in foundational models come from the University of Washington's Institute of Protein Design, run by David Baker, one of Xaira's co-founders. These models are similar to diffusion models that power image generators like OpenAI's DALL-E and Midjourney. But rather than creating art, Baker's models aim to design molecular structures that can be made in a three-dimensional, physical world."

Xaira, an AI drug discovery startup, launches with a massive $1B, says it's 'ready' to start developing drugs

#solidstatelife #ai #medicalai #drugdiscovery #chemistry

"Will AI save physicians and clinicians time and from burnout?"

"Copilots for clinicians are also becoming more common. Ambient clinical documentation is a booming business. The technology allows doctors to record conversations with patients to automatically turn them into clinical notes and summaries using AI and is a major topic at Healthcare conferences like HIMSS conference this year, where more than 30,000 health and tech professionals gathered in Orlando, Florida."

"Earlier in March, Salesforce announced Einstein Copilot: Health Actions will allow doctors to book appointments, summarize patient information and send referrals by prompting AI with conversational language."

"Administrative workloads are a major problem for clinicians across the US health-care system. A survey published (via CNBC) by Athenahealth in February found that more than 90% of physicians report feeling burned out on a regular basis, largely because of the paperwork they are expected to complete."

"I used to be part of an admissions committee for a medical school. When I interviewed idealistic young people applying to medical school, 'typing' and 'filling out forms' was never once mentioned as a reason for becoming a physician."

She goes on to describe using AI for prior authorization letters that have to be written to insurance companies. These require a letter to be written to justify the use of a drug or therapy for a specific patient and to contain details of that specific patient and why that patient needs that therapy. These are frequently rejected by the insurance companies and have to be re-written over and over to eventually get approval. "A third of medical offices employ full-time staff to take care of the average 30 prior authorizations per physician per week."

On the flip side, "the insurers have started to use AI to deny claims more quickly."

Another use is referral letters from one physician to another. "Like prior authorization letters, these are pretty formulaic."

But the thing she has the most enthusiasm for is what she calls "ambient scribes". "Ambient scribes" are AI systems that listen in to the conversation between the patient and the physician and create a templated note for the medical record. "This technology allows physicians to avoid looking at a screen and typing while they're trying to connect with a patient."

"I've tried versions from multiple AI scribe companies (including TORTUS AI, which - full disclosure - I consult for) and they do an amazing job of filtering out irrelevant information and putting the information in the right spot."

"Think of the technological challenge inherent in this process: patient visits are often interrupted by clinic staff or phone calls, meander off into conversations about kids and dogs, and use abbreviations and technical jargon. They're often circular, meaning a patient will mention a symptom and the physician won't ask a follow up question about it until several minutes later. These tools produce a full transcript that uses generative AI to find the important information and put it into a form that's indistinguishable from what a physician would actually type. Many of my friends have reported that ambient scribes actually do a better job of including important details than they would have included themselves."

Will AI save physicians and clinicians time and from burnout?

#solidstatelife #ai #voicetotext #nlp #genai #llms #medicalai

"US FDA clears DermaSensor's AI-powered skin cancer detecting device."

"The Miami-based private company's device -- also called DermaSensor -- utilizes light and an AI-powered algorithm to help primary care physicians in identifying the presence of cancer in suspicious moles or lesions."

US FDA clears DermaSensor's AI-powered skin cancer detecting device

#solidstatelife #ai #medicalai

"Each day, around 350 people in the United States die from lung cancer. Many of those deaths could be prevented by screening with low-dose computed tomography (CT) scans. But scanning millions of people would produce millions of images, and there aren't enough radiologists to do the work. Even if there were, specialists regularly disagree about whether images show cancer or not. The 2017 Kaggle Data Science Bowl set out to test whether machine-learning algorithms could fill the gap."

"The Data Science Bowl provided chest CT scans from 1,397 patients to hundreds of teams, for the teams to develop and test their algorithms. At least five of the winning models demonstrated accuracy exceeding 90% at detecting lung nodules. But to be clinically useful, those algorithms would have to perform equally well on multiple data sets."

"To test that, Kun-Hsing Yu, a data scientist at Harvard Medical School in Boston, Massachusetts, acquired the ten best-performing algorithms and challenged them on a subset of the data used in the original competition. On these data, the algorithms topped out at 60-70% accuracy. In some cases, they were effectively coin tosses." "Almost all of these award-winning models failed miserably."

The reproducibility issues that haunt health-care AI

#solidstatelife #ai #medicalai #reproducibility

"How a largely untested AI algorithm crept into hundreds of hospitals". "The use of algorithms to support clinical decision-making isn't new. But historically, these tools have been put into use only after a rigorous peer review of the raw data and statistical analyses used to develop them. Epic's Deterioration Index, on the other hand, remains proprietary despite its widespread deployment. Although physicians are provided with a list of the variables used to calculate the index and a rough estimate of each variable's impact on the score, we aren't allowed under the hood to evaluate the raw data and calculations.

"Furthermore, the Deterioration Index was not independently validated or peer-reviewed before the tool was rapidly deployed to America's largest healthcare systems."

How a largely untested AI algorithm crept into hundreds of hospitals

#solidstatelife #ai #medicalai #deteriorationindex