hideharu miura’s Post

Researchers have developed a flexible AI model, referred to as "ChatGPT-like," capable of performing a variety of diagnostic tasks across 19 cancer types, including detecting cancer cells, identifying tumor origins, and predicting genetic mutations linked to treatment responses. This model, called Clinical Histopathology Imaging Evaluation Foundation (CHIEF), was trained on millions of pathology images using self-supervised learning, allowing it to recognize patterns and adapt to a broad range of tissues. Tested on over 32 independent datasets from 24 hospitals, CHIEF has outperformed prior AI systems in cancer diagnosis and prognosis. While promising, further development is needed, especially in rare cancers and pre-cancerous conditions, to ensure its clinical application and improve outcomes for patients globally. Read more https://lnkd.in/gMjdUDGx

AI path News,Nov19,2024. New AI Method Measures Cancer Severity Using Pathology Reports. Cedars-Sinai investigators use an AI model to automate cancer pathology reports, accelerating the pace of research and potentially expanding patients’ access to clinical trials. https://lnkd.in/dZZBAnNC #AI #transformermodel #LLM #cancer #pathology #TNMstaging

🚨 𝗚͟𝗮͟𝗺͟𝗲͟𝗰͟𝗵͟𝗮͟𝗻͟𝗴͟𝗲͟𝗿͟ ͟𝗜͟𝗻͟ ͟𝗠͟𝗲͟𝗱͟𝗶͟𝗰͟𝗮͟𝗹͟ ͟𝗗͟𝗶͟𝗮͟𝗴͟𝗻͟𝗼͟𝘀͟𝘁͟𝗶͟𝗰͟𝘀͟ ͟𝗮͟𝗻͟𝗱͟ ͟𝗜͟𝗺͟𝗮͟𝗴͟𝗶͟𝗻͟𝗴͟! 🚨 𝗔𝗜 𝗖𝗼𝘂𝗹𝗱 𝗦𝗹𝗮𝘀𝗵 𝗨𝗻𝗻𝗲𝗰𝗲𝘀𝘀𝗮𝗿𝘆 𝗣𝗿𝗼𝘀𝘁𝗮𝘁𝗲 𝗕𝗶𝗼𝗽𝘀𝗶𝗲𝘀 A recent study suggests AI could significantly reduce unnecessary prostate biopsies. This is H͟U͟G͟E for both patients and healthcare systems. Here's the breakdown: 𝗧𝗵𝗲 𝗜𝘀𝘀𝘂𝗲: Prostate biopsies are invasive and can cause discomfort. However, they're crucial for definitive cancer diagnosis. 𝗧𝗵𝗲 𝗜𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻:  Researchers developed an AI algorithm that analyses MRI scans to more accurately assess prostate cancer risk. This could identify men who don't need biopsies, sparing them unnecessary procedures. 𝗧𝗵𝗲 𝗶𝗺𝗽𝗮𝗰𝘁: Reduced patient anxiety and physical discomfort. More efficient use of healthcare resources. Potential cost savings for healthcare systems. 𝗪𝗵𝗼'𝘀 𝘁𝗼 𝗹𝗼𝗼𝗸 𝗼𝘂𝘁 𝗳𝗼𝗿? Companies like Paige, specialising in AI powered diagnostic imaging and iCAD, known for their computer-aided detection (CAD) systems are a couple of companies at the forefront of AI in medical imaging. Larger players such as Siemens Healthineers and GE HealthCare have continued to invest in these areas. #𝗔𝗜 #𝗖𝗮𝗻𝗰𝗲𝗿𝘀𝗰𝗿𝗲𝗲𝗻𝗶𝗻𝗴 #𝗗𝗶𝗮𝗴𝗻𝗼𝘀𝘁𝗶𝗰𝗜𝗺𝗮𝗴𝗶𝗻𝗴 #𝗣𝗿𝗼𝘀𝘁𝗮𝘁𝗲𝗖𝗮𝗻𝗰𝗲𝗿 https://lnkd.in/e3q5AwFD

"A total of 52 breast cancer cell lines, 25 anchor drugs, and 51 library drugs were considered in developing the QSAR model. It was observed that Deep Neural Networks (DNNs) achieved an impressive R2 (Coefficient of Determination) of 0.94, with an RMSE (Root Mean Square Error) value of 0.255, making it the most effective algorithm for developing a structure-activity relationship with strong generalization capabilities. In conclusion, applying combinational therapy alongside ML and DL techniques represents a promising approach to combating breast cancer." #cancerresearch #machinelearning #deeplearning #algorithmdevelopment #neuralnetworks https://lnkd.in/emAzWmtZ

Emerging clinical applications for AI that could revolutionise patient care. Researchers from Harvard Medical School have developed a game-changing Chat-GPT like model to facilitate and automate the histological diagnosis of cancer. This type of emerging AI technology could have a significant impact on Sponsors like us developing targeted cancer therapies. Exciting times! The model was recently published in Nature: Wang, X., Zhao, J., Marostica, E. et al. A pathology foundation model for cancer diagnosis and prognosis prediction. Nature (2024).

An AI model for improved prostate cancer diagnosis Researchers have developed an artificial intelligence model that can diagnose prostate cancer better than hospital doctors. Experts at Redbud University in the Netherlands have created a computer system that has been trained and tested on more than 10,000 prostate MRI examinations. The study found that half as many ‘false positive’ results were produced using artificial intelligence. Doctors believe the system could help reduce the number of asymptomatic diagnoses and unnecessary surgeries in the most common cancer affecting men. According to the researchers, scan inspection using AI will be crucial in meeting the growing demand for medical imaging worldwide. In a first-of-its-kind study, an international team of researchers developed the technology using 10,207 MRI scans of 9,129 patients in the Netherlands. It was later tested on the scans of another 1,000 patients to determine whether these people had cancer and, if so, how severe the disease was. The technology was compared with 60 radiologists from 20 countries with an average of five to 10 years of experience performing MRI scans. #model #improved #prostate #cancer #diagnosis

An AI model for improved prostate cancer diagnosis Researchers have developed an artificial intelligence model that can diagnose prostate cancer better than hospital doctors. Experts at Redbud University in the Netherlands have created a computer system that has been trained and tested on more than 10,000 prostate MRI examinations. The study found that half as many ‘false positive’ results were produced using artificial intelligence. Doctors believe the system could help reduce the number of asymptomatic diagnoses and unnecessary surgeries in the most common cancer affecting men. According to the researchers, scan inspection using AI will be crucial in meeting the growing demand for medical imaging worldwide. In a first-of-its-kind study, an international team of researchers developed the technology using 10,207 MRI scans of 9,129 patients in the Netherlands. It was later tested on the scans of another 1,000 patients to determine whether these people had cancer and, if so, how severe the disease was. The technology was compared with 60 radiologists from 20 countries with an average of five to 10 years of experience performing MRI scans. #model #improved #prostate #cancer #diagnosis

Sex-Based Differences in Brain Cancer Risks Revealed by AI Researchers developed an AI model to identify sex-specific risk factors in aggressive brain cancer, glioblastoma. Using digital pathology slides, the model detects subtle patterns in tumors that correlate with patient survival times and differ by sex. For instance, tumors in females that infiltrate healthy tissue signal higher risk, while certain cells in male tumors are linked to faster progression. This breakthrough could pave the way for more personalized treatment plans, enhancing the quality of life for glioblastoma patients. Key Facts 1. AI identifies sex-specific risk patterns in aggressive brain tumors. 2. Infiltrating tumors in women and specific cells in men correlate with higher risk. 3. Findings could lead to personalized, targeted glioblastoma treatments. Source: https://lnkd.in/gpP2Yddm

Summary of five recent articles on AI: • Feature selection methods significantly impact the performance of Radiomic models for diagnosing clinically significant prostate cancer (csPCa). • ADC-derived Radiomic features yield more robust models than those derived from T2-weighted images. • An AI model outperformed 62 radiologists in detecting csPCa on MRI, achieving both statistical non-inferiority and superiority. • Patients favor AI involvement alongside radiologists in prostate cancer diagnosis, accepting autonomous AI dependent on its performance. • Prostate MRI and AI demonstrate promising diagnostic accuracy for localization, follow-up, and risk estimation in Active Surveillance (AS) cohorts. • AI-based integration of prostate MRI shows promise for Active Surveillance (AS) but lacks sufficient evidence for implementation. • Standardized studies with consistent inclusion criteria and progression definitions are needed. My comment: AI has primarily been tested in retrospective studies with selected patients. Readers faced limitations due to unfamiliar workstations and scanners. Most AI tools used were academic prototypes, not commercial products.

Article Sharing 👇 Title: The current use of artificial intelligence in testicular cancer: a systematic review Authors: Chuluunbaatar, Yanjinlkham, Saakshi Bansal, Andrew Brodie, Anand Sharma, Nikhil Vasdev DOI: 10.20517/ais.2023.26 How to cite: Chuluunbaatar Y, Bansal S, Brodie A, Sharma A, Vasdev N. The current use of artificial intelligence in testicular cancer: a systematic review. Art Int Surg 2023;3:195-206. https://lnkd.in/gCtfGmTX Keywords: #ArtificialIntelligence #deeplearning #machinelearning #testicularcancer #oncology