artificial intelligence Archives - Amazing Health Advances

MIT Study Warns Regular ChatGPT Use Erodes Critical Thinking, Creates “Cognitive Bankruptcy”

The AHA! Team — Mon, 04 Aug 2025 05:04:47 +0000

Lance D Johnson via Natural News – The MIT study exposes a troubling paradox: while AI promises to democratize learning, it may also stunt intellectual development. In an era where artificial intelligence promises to revolutionize education, a groundbreaking MIT study delivers a sobering reality check: reliance on AI tools like ChatGPT may be crippling the next generation’s ability to think independently. As schools rush to integrate large language models (LLMs) into classrooms, researchers warn that these systems are not just assisting students—they’re replacing the very cognitive processes essential for deep learning, problem-solving, and intellectual growth. The study, conducted by MIT’s Media Lab, reveals that students using ChatGPT for essay writing exhibited alarmingly low brain activity, weak memory retention, and diminished ownership of their work compared to those who relied on traditional research or their own knowledge. Key points: MIT researchers found ChatGPT users showed the lowest neural engagement and produced the weakest essays in quality, coherence, and originality. Brain scans (EEG) confirmed widespread cognitive disengagement—AI users copied and pasted text with minimal critical analysis. Google searchers performed moderately, while the “brain-only” group demonstrated the highest cognitive activation and retention. Lead researcher Nataliya Kosmyna warns policymakers against “GPT kindergarten”, fearing irreversible damage to developing minds. AI’s convenience comes at a cost: passive consumption replaces active learning, eroding problem-solving skills and intellectual autonomy. The cognitive cost of AI dependency The study divided participants into three groups: one using ChatGPT, another using Google, and a third relying solely on their own knowledge to write SAT-style essays. EEG monitoring revealed stark differences in brain activity. ChatGPT users displayed scattered, shallow neural patterns, suggesting their minds were on autopilot—processing information superficially without deep synthesis. In contrast, the brain-only group showed intense, coordinated activation across regions tied to critical thinking, memory, and creativity. “What really motivated me to put it out now before waiting for a full peer review is that I am afraid in 6-8 months, there will be some policymaker who decides, ‘let’s do GPT kindergarten,’” Kosmyna told TIME. “I think that would be absolutely bad and detrimental. Developing brains are at the highest risk.” The findings align with growing concerns about “cognitive offloading”—the tendency to outsource mental labor to machines. Unlike traditional search engines, which require users to evaluate sources and synthesize information, ChatGPT delivers pre-packaged answers, discouraging independent analysis. Researchers noted that AI users struggled to recall their own essays days later, while brain-only participants retained detailed knowledge. Education’s dangerous AI experiment The MIT study exposes a troubling paradox: while AI promises to democratize learning, it may also stunt intellectual development. Younger users, whose brains are still forming critical neural pathways, are most vulnerable. The study’s X post reaction summarized the threat succinctly: AI isn’t boosting productivity—it’s fostering “cognitive bankruptcy.” Historical context amplifies these concerns. Decades ago, educational psychologist Lev Vygotsky emphasized that struggle is essential for growth—forcing the mind to bridge gaps in understanding builds resilience and deeper comprehension. Modern pedagogy, however, increasingly prioritizes speed and convenience over cognitive rigor. The rise of LLMs risks accelerating this decline, creating a generation fluent in regurgitating AI outputs but incapable of original thought. The path forward: Balancing tech with cognitive sovereignty Not all technology undermines learning. The study’s Google group—while outperformed by brain-only peers—still engaged in active information retrieval and evaluation, exercising decision-making skills. The key difference? Search engines demand interaction; AI tools encourage passivity. To mitigate harm, experts urge: Delaying AI integration in early education until brains mature. Structuring assignments to require analysis, not just output generation. Promoting “brain-first” learning—forcing students to grapple with ideas before seeking AI help. Developing learning methods that inspire students to seek information that is useful and to question official narratives. Using AI, not in a passive capacity, but in a way that encourages critical thinking and mastering one’s own learning experience. Utilizing AI to assist in mundane capacities that free up the mind to pursue more creative or stimulating learning endeavors that matter. As AI reshapes education, society must choose: Will we raise thinkers—or just efficient mimics of machine logic? If students are provided AI tools and taught what to think, without question or reason, then kids will grow up looking to be spoon fed narratives and generalized information. If students are provided AI tools but are taught how to think, how to question, and how to master their learning experience, then kids will be better suited to navigate the propaganda and mindlessness that AI engines could impart. Sources include: Yournews.com Scribd.com Enoch, Brighteon.ai To read the original article, click here

The post MIT Study Warns Regular ChatGPT Use Erodes Critical Thinking, Creates “Cognitive Bankruptcy” appeared first on Amazing Health Advances.

Who Gives Better Health Advice – ChatGPT or Google?

The AHA! Team — Mon, 19 May 2025 05:09:42 +0000

Dr. Chinta Sidharthan via News-Medical – Can AI chatbots like ChatGPT give better medical answers than Google? A new study shows they can — but only if you ask them the right way. How reliable are search engines and artificial intelligence (AI) chatbots when it comes to answering health-related questions? In a recent study published in NPJ Digital Medicine, Spanish researchers investigated the performance of four major search engines and seven large language models (LLMs), including ChatGPT and GPT-4, in answering 150 medical questions. The findings revealed interesting patterns in accuracy, prompt sensitivity, and retrieval-augmented model effectiveness. Large language models Some of the biggest failures by AI chatbots involved confidently giving answers that went against medical consensus, making these mistakes particularly dangerous in health settings. The internet has now become a primary source of health information The internet has now become a primary source of health information, with millions relying on search engines to find medical advice. However, search engines often return results that may be incomplete, misleading, or inaccurate. Large language models Large language models (LLMs) have emerged as alternatives to regular search engines and are capable of generating coherent answers based on vast training data. However, while recent studies have examined the performance of LLMs in specialized medical domains, such as fertility and genetics, most evaluations have focused on a single model. Additionally, there is little research comparing LLMs with traditional search engines in health-related contexts, and few studies explore how LLM performance changes under different prompting strategies or when combined with retrieved evidence. The accuracy of search engines and LLMs also depends on factors such as input phrasing, retrieval bias, and model reasoning capabilities. Moreover, despite their promise, LLMs sometimes generate misinformation, raising concerns about their reliability. Investigating LLM accuracy The present study aimed to assess the accuracy and performance of search engines and LLMs by evaluating their effectiveness in answering health-related questions and the impact of retrieval-augmented approaches. The researchers tested four major search engines The researchers tested four major search engines — Yahoo!, Bing, Google, and DuckDuckGo — and seven LLMs, including GPT-4, ChatGPT, Llama3, MedLlama3, and Flan-T5. Among these, GPT-4, ChatGPT, Llama3, and MedLlama3 generally performed best, while Flan-T5 underperformed. The evaluation involved 150 health-related binary (yes or no) questions sourced from the Text Retrieval Conference Health Misinformation Track and covered diverse medical topics. Search engines often returned top results that didn’t answer the question directly, but when they did, those answers were usually correct — highlighting a precision problem rather than accuracy. Search engines For search engines, the top 20 ranked results were analyzed. A passage extraction model was employed to identify relevant snippets, and a reading comprehension model determined whether each snippet provided a definitive answer. Additionally, user behaviors were simulated using two models: a “lazy” user who stops at the first yes or no answer and a “diligent” user who cross-references three sources before deciding. Interestingly, the study found that ‘lazy’ users achieved similar accuracy to ‘diligent’ users and, in some cases, even performed better, suggesting that top-ranked search engine results may often suffice—though this raises concerns when incorrect information ranks highly. For LLMs For LLMs, the questions were tested under different prompting conditions: no-context (just the question), non-expert (prompts were framed in the language used by laypeople), and expert (prompts were framed for guiding responses toward reputable sources). The study also tested few-shot prompts—adding a few example questions and answers to guide the model—which improved performance for some models but had limited effect on the best-performing LLMs. The study also explored retrieval-augmented generation, where LLMs were fed search engine results before generating responses. Performance Performance was assessed based on accuracy in correctly answering the questions, sensitivity to input phrasing, and improvements gained through retrieval augmentation. The researchers also used statistical significance tests to determine meaningful performance differences between models. Although some LLMs outperformed others, statistical tests showed that in many cases, performance differences between leading models were not significant, indicating that top LLMs performed comparably in many instances. Furthermore, the researchers categorized common LLM errors, such as misinterpretation, ambiguity, and contradictions with medical consensus. The study also noted that while the “expert” prompt generally guided LLMs toward more accurate responses, it sometimes increased the ambiguity of their answers. Key findings COVID-19 questions proved easier for both LLMs and search engines, likely because pandemic-related data dominated their training and indexing periods. The study found that LLMs generally outperformed search engines in answering health-related questions. While search engines correctly answered 50–70% of queries, LLMs achieved approximately 80% accuracy. However, LLM performance was highly sensitive to input phrasing, with different prompts yielding significantly varied results. The “expert” prompt, which guided LLMs toward medical consensus, was found to perform the best, although it sometimes led to less definitive answers. Among the search engines, Bing provided the most reliable results, but it was not significantly better than Google, Yahoo!, or DuckDuckGo. Moreover, many search engine results contained non-responsive or off-topic information, contributing to lower precision. However, when focusing only on responses that addressed the question, search engine precision rose to 80–90%, though about 10–15% of these still contained incorrect answers. ‘Lazy’ users Furthermore, contrary to common assumptions, the study found that ‘lazy’ users sometimes achieved similar or better accuracy with less effort, highlighting both the efficiency and the risk of trusting initial search results. Additionally, the researchers observed that retrieval-augmented methods improved LLM performance, especially for smaller models. By integrating top-ranked search engine snippets, even lightweight models such as text-davinci-002 performed similarly to GPT-4. However, the study noted that retrieval augmentation sometimes decreased performance, especially when low-quality or irrelevant search results were fed into LLMs—emphasizing the critical role of retrieval quality. For some datasets, like COVID-19-related questions from 2020, adding search engine evidence even worsened LLM performance, possibly because these questions were already well-covered in LLM training data. Feeding AI chatbots search results didn’t always help; in some cases, irrelevant or low-quality snippets actually made chatbot answers worse, showing that more information isn’t always better. Error analysis The error analysis also revealed three major failure modes for LLMs, including incorrect medical consensus understanding, misinterpretation of questions, and ambiguous answers. Notably, some health-related questions were inherently difficult, and both LLMs and search engines struggled to provide correct answers to these questions. The study also found that performance varied depending on the dataset: questions from 2020, largely focused on COVID-19, were easier for both LLMs and search engines, while the 2021 dataset presented more challenging medical questions. Overall, while LLMs demonstrated superior accuracy, their propensity to prompt variations and misinformation highlighted the need for caution in medical decision-making based on LLM answers. The study also suggested combining LLMs with search engines through retrieval augmentation could yield more reliable health answers, but only when the retrieved evidence is accurate and relevant. Conclusions In summary, the study highlighted search engines’ and LLMs’ strengths and weaknesses in answering health-related questions. While LLMs generally outperformed search engines, their accuracy was found to be highly dependent on input prompts and retrieval augmentation. Although advanced models like GPT-4 and ChatGPT performed well, other models such as Llama3 and MedLlama3 sometimes matched or even outperformed them, depending on the dataset and prompting strategy. Moreover, while combining both technologies appears promising, ensuring the reliability of retrieved information remains a challenge. The researchers emphasized that smaller LLMs when supported with high-quality search evidence, can perform on par with much larger models—raising questions about the need for ever-larger AI models when retrieval augmentation could be a viable alternative. These results suggested that future research should explore methods to enhance LLM trustworthiness and mitigate misinformation in health-related AI applications. Journal reference: Fernández-Pichel, M., Pichel, J.C. & Losada, D.E. (2025). Evaluating search engines and large language models for answering health questions. NPJ Digital Medicine. 8, 153. DOI:10.1038/s41746-025-01546-w, https://www.nature.com/articles/s41746-025-01546-w To read the original article click here.

The post Who Gives Better Health Advice – ChatGPT or Google? appeared first on Amazing Health Advances.

AI Breakthrough Slashes Celiac Diagnosis Time from Months to Minutes

The AHA! Team — Fri, 09 May 2025 05:01:21 +0000

Cassie B. via Natural News – Cambridge researchers created an AI tool diagnosing celiac disease as accurately as human pathologists but in under a minute. The AI achieved 97% accuracy in tests using 4,000+ biopsy images, reducing wait times from months to seconds. Experts highlight AI’s potential to ease NHS backlogs but note infrastructure gaps hinder adoption. Untreated celiac disease can cause severe complications, affecting 1 in 100 people globally. British researchers at the University of Cambridge have developed an artificial intelligence tool that diagnoses celiac disease with the same accuracy as human pathologists but at a fraction of the time, potentially reducing diagnosis wait times from months to less than a minute. The breakthrough, published March 27 in the New England Journal of Medicine AI, demonstrates how market-driven technological solutions could alleviate inefficiencies plaguing government-run healthcare systems like Britain’s National Health Service (NHS), where patients routinely face lengthy wait times for diagnosis and treatment. AI matches pathologist accuracy while drastically reducing wait times The machine learning algorithm was trained on more than 4,000 biopsy images from five different hospitals and tested on an independent set of 650 previously unseen images. The results showed remarkable accuracy – correctly identifying celiac disease in more than 97% of cases, with sensitivity exceeding 95% and specificity of almost 98%. “It can take many years to receive an accurate diagnosis, and at a time of intense pressures on healthcare systems, these delays are likely to continue,” said Elizabeth Soilleux, consultant hematopathologist and professor of pathology at Cambridge University, who led the research. “AI has the potential to speed up this process, allowing patients to receive a diagnosis faster, while at the same time taking pressure off NHS waiting lists.” AI model delivers results Dr. Florian Jaeckle, co-author of the research, highlighted the dramatic time savings: while human pathologists require 5-10 minutes to analyze each biopsy, the AI model delivers results “in less than a minute and as soon as a biopsy is scanned.” “Duodenal biopsies are often put at the back of the pathologist’s lists as they are not as serious as for example a possible cancer case, meaning that patients often have to wait weeks or even months to find out if they have celiac disease,” Jaeckle explained. “With AI they could get a result almost instantly… Therefore, there would never be a waiting list with AI.” Government healthcare infrastructure lags behind innovation Despite the promising technology, the president of the Royal College of Pathologists acknowledged significant barriers to implementation within Britain’s government-run healthcare system. Dr. Bernie Croal said that while the AI tool “has the potential to radically transform how we diagnose celiac disease,” the NHS lacks the necessary digital infrastructure to fully utilize such innovations. “More work will be needed to get to the point where AI is fully developed and used safely in the NHS,” Croal admitted. “Investment in digital pathology, joined up functional IT systems… as well as training for pathologists to understand and use AI, will all need to be put in place.” These infrastructure shortcomings highlight a persistent pattern in government-managed healthcare: while private sector innovation rapidly advances diagnostic and treatment capabilities, bureaucratic systems struggle to keep pace with technological progress. Celiac disease affects approximately one in 100 people, causing symptoms including stomach cramps, diarrhea, skin rashes, weight loss, fatigue, and anemia when patients consume gluten. When left untreated, it can lead to serious complications including malnutrition, osteoporosis, infertility, and increased risk of certain cancers. The Cambridge researchers have established a spinout company, Lyzeum Ltd, to commercialize the algorithm, creating a market-based pathway for this life-improving technology to reach patients while government systems catch up. The research received funding from Coeliac UK, Innovate UK, and the Cambridge Centre for Data-Driven Discovery, demonstrating how private sector partnerships can accelerate medical breakthroughs without total reliance on government resources. Sources for this article include: TheGuardian.com Cam.ac.uk MedicalXpress.com To read the original article, click here

The post AI Breakthrough Slashes Celiac Diagnosis Time from Months to Minutes appeared first on Amazing Health Advances.

Can AI Recognize the Signs of Depression in People’s Voices?

The AHA! Team — Fri, 28 Mar 2025 05:07:05 +0000

Dr. Chinta Sidharthan via News-Medical – A machine learning tool successfully identified vocal markers of depression in over 70% of cases within 25 seconds, highlighting its potential for improving mental health screening in primary care and virtual healthcare settings. In a recent article in The Annals of Family Medicine, researchers evaluated the effectiveness of a machine learning (ML) tool for detecting vocal signs linked to severe or moderate depression. The tool successfully detected vocal markers of depression in just 25 seconds, correctly identifying cases of depression in more than 70% of samples, highlighting its utility for mental health screening. Background Depression is a major health issue, affecting about 18 million Americans annually, with nearly 30% experiencing it at some point in their lives. Despite guidelines recommending universal screening, depression screening in primary care remains very low (

The post Can AI Recognize the Signs of Depression in People’s Voices? appeared first on Amazing Health Advances.

The AI Tech That Can Spot Serious Illness Before the Doctor

The AHA! Team — Wed, 05 Mar 2025 06:16:00 +0000

John Jeffay via Israel21c – Lavaa Health’s platform can identify disease and hard-to-diagnose illnesses at the earliest stage, allowing doctors to quickly draw up efficient treatment plans. Meet your GP’s new best friend – artificial intelligence (AI). Lavaa Health, an Israeli startup, watches over all patient data, ready to spot early signs of potential health issues, and uses its vast medical database to identify hard-to-diagnose or rare illnesses. It’s a virtual assistant that works in the background to offer help and alerts, but leaves the physician very much in the driver’s seat, from making the diagnosis to drawing up a treatment plan. The company was founded after a family tragedy. Adam Amitai, Lavaa’s CEO, watched helplessly as his 55-year-old mother-in-law succumbed to ovarian cancer. It had taken a year for the doctors to correctly diagnose her, by which time it was too late. She died eight months later. Amitai doesn’t blame the physicians and says they provided excellent care. But he realized they weren’t exploiting the power of AI to get quicker and more accurate insights. And so he interviewed 200 physicians in the United States, to fully understand how AI could best help them. And he drew on his seven years’ experience as an “offensive cyber officer” in the IDF – where a key challenge was sifting vital details from masses of data. Amitai had also continued to work in intelligence afterwards and had set up an automated trading platform for institutional investors. So, he wasn’t from the world of healthcare, but he recognized that it could benefit from advanced systems that had been developed elsewhere. Handling data more efficiently “I understood there was a big problem with data handling in the healthcare industry,” he tells ISRAEL21c. He saw it when each of his three children were born. Every time, the doctor asked for the family’s medical history. And he saw it with the death of his mother-in-law. He believes AI would have suggested ovarian cancer as a diagnosis much sooner. The main problem is in primary care, the people who are in charge of your health on a daily basis “It’s not the physician’s fault, it’s not the care team fault, they’re doing their best, but they just don’t have the tools,” he says. “The main problem is in primary care, the people who are in charge of your health on a daily basis. They’re reactive instead of proactive. They’re trying to solve a single problem, not your whole health.” And they generally lack the resources to understand what the problem is and to diagnose it correctly. Lavaa’s AI-powered Preventive Care Engine Platform assists the physician by offering evidence-based insights. “We are not allowing the computer to try to automatically detect the conditions. We’re using the accepted worldwide care protocols, but we’re using AI to extract the data,” says Amitai. “Physicians cannot go through all of this data by themselves in the amount of time that they have. It’s just impossible, so this is giving them a huge backup. “The number of parameters for a physician to check and the number of possible diseases is infinite, and time is limited. But computers are really good at matching parameters to diseases. “I realized that technology from the intelligence world already did this, so it was a question of applying it to healthcare.” Prevention, intervention Lavaa is all about prevention and early intervention. Its AI platform can generate questions for a particular patient based on what it sees in their records. It may, for example, ask if a female patient remembers the age at which she had her first period – something that’s relevant for breast cancer, but is never recorded in an EMR (electronic medical record). Or it may send targeted messages, questionnaires, or notifications. It acts as an early warning system, designed to prevent the development of chronic or psychological diseases, and cancer. Lavaa currently looks after over 700,000 patients, all in the US, though the company has plans to expand globally. Amitai estimates the technology has so far saved 1,500 lives. “These are people who had a condition that could have been terminal but caught it on time and we managed to alert the physician, which meant the patients got either the right or better drugs, and better treatment, or a referral to the right place,” he says. Lavaa is not the only such AI solution, but Amitai says the healthcare market is big enough for everybody. Some other companies use AI to both inform and to diagnose – unlike Lavaa – or as a “black box” providing a diagnosis but no explanation of its “thinking.” The company has 12 staff members at its offices in Ra’anana, central Israel, and a team working in the US. Lavaa was founded in 2021, has attracted $5 million in investments. A Series A funding round will be launched later this year. “We want to go global,” Amitai says. “Our solution can work anywhere, and we believe it can improve healthcare around the world.” For more information, click here. To read the original article click here.

The post The AI Tech That Can Spot Serious Illness Before the Doctor appeared first on Amazing Health Advances.

New Startup Is Training Labradors to Detect Cancer

AHA Publisher — Wed, 31 Aug 2022 07:00:48 +0000

Abigail Klein Leichman via Israel21c – Few people better understand a dog’s psyche and capabilities than Col. (Res.) Ariel Ben-Dayan, former commander of the Israel Defense Forces’ Oketz canine unit. Ben-Dayan is CEO of SpotitEarly, which is building an early cancer detection system based on dogs’ superior sense of smell combined with artificial intelligence. The user will purchase a home kit containing a surgical mask, breath into the mask for five minutes and then send it to a screening lab where the mask is inserted into a sniffing station. Several specially trained dogs will sniff the sample. Few people better understand a dog’s psyche and capabilities than Col. (Res.) Ariel Ben-Dayan, former commander of the Israel Defense Forces’ Oketz canine unit. Ben-Dayan is CEO of SpotitEarly, which is building an early cancer detection system based on dogs’ superior sense of smell combined with artificial intelligence. The user will purchase a home kit containing a surgical mask, breath into the mask for five minutes and then send it to a screening lab where the mask is inserted into a sniffing station. Several specially trained dogs will sniff the sample. The company’s lab is on a kibbutz, where the dogs live and work. SpotitEarly is conducting clinical trials with two Israeli medical centers, Hadassah in Jerusalem and Sourasky in Tel Aviv. “During the pre-trial training phase, our test showed higher sensitivity and better specificity in detecting early stages of cancer compared to any existing test on the market today,” says Ben-Dayan. “We expect interim analysis in three months and will finish clinical trials in 10 months.” Dog and Nanotechnology Hybrid Scent technology is a hot new frontier, and there are disease-sniffing electronic “noses” being developed at Ben-Gurion University and at the Technion-Israel Institute of Technology. However, e-noses have not yet proved accurate enough in real-world conditions where it’s necessary to isolate disease odors from other scents in a sample – such as coffee on the patient’s breath. Ben-Dayan believes the future lies in a hybrid electronic nose and dog nose. “I have many years of experience with dogs, and I know that no one can replace a dog nose. But when you add technology, this will be the optimal system to give us another level of confidence and verification,” he tells ISRAEL21c. “If we find cancer, we will then send our customer for additional screening procedures. If those scans don’t find anything [yet] because it’s too early stage, we will recommend routine screenings every six or 12 months, since we assume there are cases where we’ll find it before any other lab,” Ben-Dayan says. He and his cofounders — Roi Ophir, Udi Bobrovsky and Ohad Sharon — recently added three PhDs to their team, one an expert in data science, one in animal behavior and the other a clinical lab manager. They have high hopes that eventually their canine and AI system could screen for multiple types of cancer. Scalability But is it possible to scale a business based on live animals? Ben-Dayan believes the kits make it doable. And judging by the $6.2 million seed round SpotitEarly recently raised, investors such as Hanaco Ventures also believe the idea is feasible. Ben-Dayan tells ISRAEL21c that one lab could perform a million tests annually. “We are building an automatic lab so operational efforts will be minimal. The dogs will be able to check hundreds of screening kits in an hour,” says Ben-Dayan, emphasizing that this task is like a fun game for them. He envisions each SpotitEarly lab situated in a dog-friendly setting where the canines would live on premises and have “the best lodgings, food, veterinary care, love and care possible” as they do at the test lab in Israel. Though he and his staff are dog lovers, he says, the company is founded on their concern for human life. Given that many people are reluctant to go for tests such as mammography and colonoscopy, this is an easy, relatively inexpensive alternative people could perform at home. “Our mission is increasing the number of people who do screening tests,” says Ben-Dayna. “I’ll be very happy to sell a million kits so we can save the lives of 5,000 people – that’s the number we estimate according to statistics.” If their behavior, confirmed by an AI algorithm, indicates that cancer is detected, the person will be referred to the medical system for further testing. According to the American Kennel Club, dogs’ ultrasensitive noses can detect the odor of cancer in breath or urine at a very early stage, when the disease is far more treatable. SpotitEarly trains Labradors to sniff out breast, lung, colon or prostate cancer reliably from one sample. Ben-Dayan says this breed has the right set of traits: excellent smelling capability, discipline and passion to work. To read the original article click here.

The post New Startup Is Training Labradors to Detect Cancer appeared first on Amazing Health Advances.

A Blood Test That Can Identify Recurrent Cancer

AHA Publisher — Fri, 15 Oct 2021 07:00:04 +0000

Abigail Klein Leichman via Israel21c – When Asaf Zviran was diagnosed with cancer, he was doing operations research and R&D in the Israeli navy and earning his master’s degree at the Technion-Israel Institute of Technology. Maybe that’s why he pictured cancer as an enemy to defeat– if he could find the right weapon and target. After his successful treatment and seven years of military service, in 2012Zviranbegan PhD studies in molecular biology at the Weizmann Institute of Science. During his postdoctoral research at the New York Genome Center from 2016 to 2019, he zeroed in on his target: cancer that persists or returns after treatment. Working with world-class scientists at the center, he invented a whole-genome sequencing method, aided by artificial intelligence, enabling early detection of persistent or recurrent cancer cells from a standard blood sample. As soon as he finished his postdoc fellowship, Zviran and three friends founded C2i Genomics to develop his personalized medicine approach, described in a paper published in Nature Medicine. His postdoctoral mentor, Weill Cornell Medicine oncologist Dan Landau, is the company’s scientific cofounder and sits on its scientific advisory board. The company has raised more than $100 million in financing from Casdin Capital, NFX, Duquesne Family Office, Section 32, iGlobe Partners, Driehaus Capital and others. “With our technology, physicians can monitor their patient treatment response and detect treatment failure or disease recurrence months and even years before they would do otherwise,” Zviran says. Command, Control and Intelligence “C2i is a military term for ‘command, control and intelligence’ – it expresses the vision of applying defense methodologies to oncology,” explains cofounder Boris Oklander, the startup’s CTO. “Cancer is like an enemy with a unique signal of a mutation that we want to detect. We fight this enemy with technology grounded in our experience in the defense sector.” Headquartered in New York with R&D in Haifa and a sequencing lab in Cambridge, Massachusetts, C2i Genomics is one of nine digital healthcare startups in the current cohort of PlayBeyondBio. This is an accelerator run by a partnership of Israeli venture capital fund JVP, British-Swedish pharma giant AstraZeneca, international consulting firm Accenture, Margalit Startup City, Amazon AWS, and Shaare Zedek Medical Center in Jerusalem. “The idea is to see how this ecosystem of partners can join forces and generate pilot studies,” says Oklander. “There are biobanks of patient samples in Shaare Zedek that can help evaluate and validate technologies like ours and help get our plug-and-play diagnostic service to market quickly.” In July, C2i signed a collaboration agreement with Premier, an American healthcare improvement company. The agreement includes implementing the C2i platform at eight Premier member hospitals and clinics. Ending Over- and Undertreatment Oklander tells ISRAEL21c that this unique approach could solve the significant problem of over- or undertreatment of solid tumors. What usually happens is that the tumor is removed surgically. The pathology report helps the oncologist make an educated guess whether to wait and monitor the patient periodically, or to start chemotherapy and/or radiation as a precaution against undetected cancer cells. “There is no good way to know in real time if the patient is cancer-free,” says Oklander. Patients getting chemotherapy need to wait months before knowing if there is a good response. Patients getting monitored may have a tumor growing undetected during that time. “If we can measure the level of cancer in the patient’s blood in real time — like measuring the glucose level in diabetics — there is no need to predict but rather measure what is happening right now and support clinical decision-making relating to which treatment to use, or to monitor and step in as soon as necessary.” With 20 million new cancer diagnoses around the world each year, this method could save many people from being overtreated with toxic, painful, costly chemotherapy and from being undertreated while a new tumor is quietly growing. “Our test can improve the entire cycle for the patient and for the payers,” says Oklander, noting that insurance carriers are shifting their focus to evidence-based reimbursement to improve the economics of oncology. World First Although other cancer genomics services are available, C2i’s platform is designed to be uniquely accessible and thorough. “The way our competitors work is that patient blood samples are sent to a central lab, usually in the United States,” says Oklander. “Our distributed solution is logistically simpler because the sample can be processed at any genomic sequencing lab. There are 15,000 sequencers in the world — the machines that produce genomic data from blood samples. Each of these sequencers can communicate with us through a secure Internet connection.” Compliance with GDPR (Europe) and HIPAA (USA) privacy regulations is configured into the C2i system for each region, he adds. The other difference is that C2iscans the entire genome to find thousands, or hundreds of thousands, of mutations unique to each patient. “In two people with the same cancer diagnosis, less than 1% of the mutations are common between the two of them,” explains Oklander. “We capture the entire genomic makeup of each patient using AI and signal processing. Then we test the blood at all stages in the treatment to see if these mutations are still present.” Zviran notes that C2i Genomics was a finalist for the 2021 Spinoff Prize awarded by Nature and Merck. “We are proud of the recognition of our work detecting ultra-low amounts of tumor DNA in blood samples, to guide treatment decisions and ultimately save patient lives,” Zviran says. For more information, click here To read the original article click here.

The post A Blood Test That Can Identify Recurrent Cancer appeared first on Amazing Health Advances.

Artificial Intelligence Could Crack the Language of Cancer and Alzheimer’s

AHA Publisher — Fri, 16 Apr 2021 07:00:03 +0000

St. John’s College, University of Cambridge via EurekAlert – Powerful algorithms used by Netflix, Amazon and Facebook can ‘predict’ the biological language of cancer and neurodegenerative diseases like Alzheimer’s, scientists have found. Big data produced during decades of research was fed into a computer language model to see if artificial intelligence can make more advanced discoveries than humans. Academics based at St John’s College, University of Cambridge, found the machine-learning technology could decipher the ‘biological language’ of cancer, Alzheimer’s, and other neurodegenerative diseases. Their ground-breaking study has been published in the scientific journal PNAS today (April 8 2021) and could be used in the future to ‘correct the grammatical mistakes inside cells that cause disease’. Professor Tuomas Knowles, lead author of the paper and a Fellow at St John’s College, said: “Bringing machine-learning technology into research into neurodegenerative diseases and cancer is an absolute game-changer. Ultimately, the aim will be to use artificial intelligence to develop targeted drugs to dramatically ease symptoms or to prevent dementia happening at all.” Every time Netflix recommends a series to watch or Facebook suggests someone to befriend, the platforms are using powerful machine-learning algorithms to make highly educated guesses about what people will do next. Voice assistants like Alexa and Siri can even recognise individual people and instantly ‘talk’ back to you. Dr Kadi Liis Saar, first author of the paper and a Research Fellow at St John’s College, used similar machine-learning technology to train a large-scale language model to look at what happens when something goes wrong with proteins inside the body to cause disease. She said: “The human body is home to thousands and thousands of proteins and scientists don’t yet know the function of many of them. We asked a neural network based language model to learn the language of proteins. “We specifically asked the program to learn the language of shapeshifting biomolecular condensates – droplets of proteins found in cells – that scientists really need to understand to crack the language of biological function and malfunction that cause cancer and neurodegenerative diseases like Alzheimer’s. We found it could learn, without being explicitly told, what scientists have already discovered about the language of proteins over decades of research.” Proteins are large, complex molecules that play many critical roles in the body. They do most of the work in cells and are required for the structure, function and regulation of the body’s tissues and organs – antibodies, for example, are a protein that function to protect the body. Alzheimer’s, Parkinson’s and Huntington’s diseases are three of the most common neurodegenerative diseases, but scientists believe there are several hundred. In Alzheimer’s disease, which affects 50 million people worldwide, proteins go rogue, form clumps and kill healthy nerve cells. A healthy brain has a quality control system that effectively disposes of these potentially dangerous masses of proteins, known as aggregates. Scientists now think that some disordered proteins also form liquid-like droplets of proteins called condensates that don’t have a membrane and merge freely with each other. Unlike protein aggregates which are irreversible, protein condensates can form and reform and are often compared to blobs of shapeshifting wax in lava lamps. Professor Knowles said: “Protein condensates have recently attracted a lot of attention in the scientific world because they control key events in the cell such as gene expression – how our DNA is converted into proteins – and protein synthesis – how the cells make proteins. “Any defects connected with these protein droplets can lead to diseases such as cancer. This is why bringing natural language processing technology into research into the molecular origins of protein malfunction is vital if we want to be able to correct the grammatical mistakes inside cells that cause disease.” Dr Saar said: “We fed the algorithm all of data held on the known proteins so it could learn and predict the language of proteins in the same way these models learn about human language and how WhatsApp knows how to suggest words for you to use. “Then we were able ask it about the specific grammar that leads only some proteins to form condensates inside cells. It is a very challenging problem and unlocking it will help us learn the rules of the language of disease.” The machine-learning technology is developing at a rapid pace due to the growing availability of data, increased computing power, and technical advances which have created more powerful algorithms. Further use of machine-learning could transform future cancer and neurodegenerative disease research. Discoveries could be made beyond what scientists currently already know and speculate about diseases and potentially even beyond what the human brain can understand without the help of machine-learning. Dr Saar explained: “Machine-learning can be free of the limitations of what researchers think are the targets for scientific exploration and it will mean new connections will be found that we have not even conceived of yet. It is really very exciting indeed.” The network developed has now been made freely available to researchers around the world to enable advances to be worked on by more scientists. To read the original article click here.

The post Artificial Intelligence Could Crack the Language of Cancer and Alzheimer’s appeared first on Amazing Health Advances.

A New AI ‘Super Nurse’ Monitors Patients in Israeli Hospital

AHA Publisher — Sun, 08 Mar 2020 08:00:04 +0000

Brian Blum via Israel21c – Artificial intelligence can spot potential deterioration before a human nurse or doctor could, and it can predict which patients will be readmitted. Look, it’s Super Nurse! Able to monitor multiple patients in separate rooms simultaneously; staying on top of their blood pressure, pulse and vital signs; and spotting signs of deterioration even before the patients feel it themselves. This medical superhero is not human, but rather a product of artificial intelligence, advanced software algorithms, sensors and cameras. And it’s being assembled right now at Tel Aviv Sourasky Medical Center. The creation of an AI-powered “super nurse” is the result of a decade of steady work by Ahuva Weiss-Meilik and her team in the hospital’s I-Medata center. “Our doctors and nurses can’t be everywhere,” Weiss-Meilik tells ISRAEL21c. A thought no doubt being echoed across Chinese hospitals in the midst of the coronavirus outbreak right now. The situation is made even worse by chronic staff shortages in hospitals around the world, including those in Israel. The hospital is currently conducting a trial of Weiss-Meilik’s continuous monitoring technology with 24 patients in “regular” internal-medicine wards. They have all given their consent. (Big Brother may be watching but he has permission.) Sensors check patients’ blood pressure, heart and respiratory rate, while a camera watches at all times. The data is streamed to a central viewing station where hospital staff is alerted if a change develops. If monitoring were the entire extent of Weiss-Meilik’s project, it wouldn’t be so potentially game-changing. But the I-Medata team is developing predictive algorithms that can determine, based on the camera and sensor data (a drop in blood pressure, for example), combined with information from a patient’s electronic medical record (EMR) and aggregated data, what a particular patient’s medical trajectory is most likely to be – and whether that patient should be given extra attention. AI can spot potential deterioration hours before a human nurse or doctor could. And it can tell if a patient is likely to be readmitted in the future. “No doctor, no matter how good he or she is, would be able to look at all this data and gain immediate insight,” Weiss-Meilik points out. “AI testing will eventually be a part of all hospital admissions in the future,” she predicts. Big Hospital Data Tel Aviv Medical Center– Israel’s largest acute-care facility, with 1,500 beds — has an advantage in implementing this type of big data system because it’s long been at the forefront of collecting patient data in Israel. “We have computerized medical records going 10 years back,” Weiss-Meilik says. “It’s enabled us to do this kind of research and development.” Weiss-Meilik’s team now is creating an ecosystem within the hospital “to inform and support our clinical staff.”While the artificial intelligence and algorithm development is being kept in-house, the I-Medata department has partnered with two Israeli startups. AnyVision of Holon, which specializes in face, body and object-recognition software, is powering the cameras. The company generally works in the security and surveillance sphere, including law enforcement, retail, banking and casino clients; this is AnyVision’s first foray into powering medical technology. (The company closed a $74 million Series A financing round in June 2019.) BioBeat, based in Petah Tikva, is behind the sensors. The company has developed a disposable patch worn on the skin to track such vital signs as blood pressure, oxidation rate, pulse, skin temperature and sweat. “Doctors don’t have to measure anything,” BioBeat Chief Medical Officer Arik Eisenkraft tells ISRAEL21c. All the vitals are transmitted in real time via Bluetooth and Wi-Fi to the patient’s EMR as well as to the BioBeat cloud, so BioBeat can learn from the data and improve functionality. BioBeat has received US Food and Drug Administration clearance for its cuff-less non-invasive blood pressure monitoring. The company has also developed a wristwatch device for monitoring vital signs at home, Eisenkraft says. Making the Hospital More Intelligent Weiss-Meilik’s group is also working on integrating other areas of Israeli-developed AI into the hospital workflow. Medical imaging startups Zebra Medical and Aidoc both use artificial intelligence to help radiologists quickly spot problem areas. Agamon specializes in natural language processing (NLP) to make sense of unstructured text summaries in patients’ records, such as operation summaries and radiological reports. The system puts these summaries into a structured database format. “It takes a lot of human resources to do that,” explains Weiss-Meilik. “Using AI, we can open up data that was useless before in order to do advanced research.” Director-general Dr. Ronni Gamzu calls I-Medata’s work “the future of medicine.” While some patients are not connected to a monitor “because their condition does not require it,” Gamzu told Ynet, “sometimes they suddenly ‘surprise’ us and their condition starts to deteriorate. The medical staff needs systems to help them identify complex situations and to navigate through massive amounts of data.” Weiss-Meilik is the head of the medical center’s Data Science and Quality Division. She previously served as the national coordinator of the Middle East Quality in Health Care Program coordinated by Harvard University, and headed the clinical and economic performance unit at Sheba Medical Center. She received her PhD from the Technion – Israel Institute of Technology. “For now, we’re still at an early stage,” Weiss-Meilik tells ISRAEL21c. “But if we develop these models and get them on the market, we could improve health worldwide.” That would be a super nurse, indeed. To read the original article click here. For more articles from Israel21c click here.

The post A New AI ‘Super Nurse’ Monitors Patients in Israeli Hospital appeared first on Amazing Health Advances.