วีดีโอ

Fantastic Alec

Hi Dr Madhu Dewan from UK

AI Book: The ABC of Pathology: amzn.to/3WCcnoa

Morning

Set your scientific literature alerts like that and never get paralyzed about which papers to read next. Let me know your area of expertise below. Mine: digital pathology.

FiFine Microphone: amzn.to/3WQCTLS

As much as I love digital pathology it has both advantages and disadvantages. Let me know what annoys you most!

The abstracts we covered today: 📄 Advances in Bone Marrow Evaluation pubmed.ncbi.nlm.nih.gov/39089749/ 📄 Digital Imaging and AI Pre-classification in Hematology pubmed.ncbi.nlm.nih.gov/39089746/ 📄 Evaluation of CD3 and CD8 T-Cell Immunohistochemistry for Prognostication and Prediction of Benefit From Adjuvant Chemotherapy in Early-Stage Colorectal Cancer Within the QUASAR Trial pubmed.ncbi.nlm.nih.gov/39083705/ 📄 A Pan-Cancer Patient-Derived Xenograft Histology Image Repository with Genomic and Pathologic Annotations Enables Deep Learning Analysis A survey analysis of the adoption of large language models among pathologists pubmed.ncbi.nlm.nih.gov/39082680/ 📄 Clinical-Grade Validation of an Autofluorescence Virtual Staining System with Human Experts and a Deep Learning System for Prostate Cancer pubmed.ncbi.nlm.nih.gov/39069201/

I love that I want to see more

Too many commercials1

The abstracts we discussed today: 1) A visual-language foundation model for computational pathology www.nature.com/articles/s41591-024-02856-4 2) A foundation model for clinical-grade computational pathology and rare cancers detection www.nature.com/articles/s41591-024-03141-0 The tool we tried today: Undermind: 2ly.link/1yjS6 A book I recommend: "ABC of Artificial Intelligence. Baby University": amzn.to/4fyaXE7

Hi Dr Bindu from India

❤

Are these microscope cameras allowed for remote primary diagnosis?

C

The dataset the AI itself is trained with matters immensly though - this is how it synthesizes examples. I wouldnt trust any AI tool, in any situation, to create larger sample sets than the datasets it is referencing, and have those samples still conform to what you might encounter in the real world. And if the AI is trained off even a smidge of the internet, for it's language model even, then it will be stitching together language from an uncountable number of untrustable sources (couch doctors etc.) without any way to distinguish credibility. Despite all the frenzy for AI, when it comes to training medical professionals, I can't agree that teaching through AI synthesized examples is better than teaching by experience. In my eyes, this is equivalent to turning medical training into an assembly line - less quality teaching, for a higher output of less qualified medical professionals.

AI has such potential! Thanks for sharing your thoughts.

Here are the abstracts we reviewed today: Improving the diagnostic performance of inexperienced readers for thyroid nodules through digital self-learning and artificial intelligence assistance www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2024.1372397/full U.S. payer budget impact of using an AI-augmented cancer risk discrimination digital histopathology platform to identify high-risk of recurrence in women with early-stage invasive breast cancer www.tandfonline.com/doi/full/10.1080/13696998.2024.2379211 Evaluating the efficacy of few-shot learning for GPT-4Vision in neurodegenerative disease histopathology: A comparative analysis with convolutional neural network model onlinelibrary.wiley.com/doi/10.1111/nan.12997 ChatGPT for digital pathology research pubmed.ncbi.nlm.nih.gov/38987117/ Evaluating artificial intelligence-enhanced digital urine cytology for bladder cancer diagnosis acsjournals.onlinelibrary.wiley.com/doi/10.1002/cncy.22884

Would you prefer working remotely or on site? why? Let me know below.

Thank you for sharing. I am also in the same industry, but there is very little exchange of relevant information. This channel is very precious.

The abstracts reviewed today: 1) Latent Diffusion Models with Image-Derived Annotations for Enhanced AI-Assisted Cancer Diagnosis in Histopathology. pubmed.ncbi.nlm.nih.gov/39001331/ 2)Deep Learning-Based Dynamic Risk Prediction of Venous Thromboembolism for Patients With Ovarian Cancer in Real-World Settings From Electronic Health Records. pubmed.ncbi.nlm.nih.gov/38996199/ 3) Deep learning empowered breast cancer diagnosis: Advancements in detection and classification. pubmed.ncbi.nlm.nih.gov/38990817/ 4) ChatGPT for digital pathology research pubmed.ncbi.nlm.nih.gov/38987117/ 5) Joining forces for pathology diagnostics with AI assistance: The EMPAIA initiative pubmed.ncbi.nlm.nih.gov/38984198/ 6) Possible benefits, challenges, pitfalls, and future perspective of using ChatGPT in pathology pubmed.ncbi.nlm.nih.gov/38971620/

Hi, I came from Thailand, but I'm now a final-year PhD student in Japan! I am really interested in digital pathology, and your book has helped me a lot! I look forward to your next videos.

If you ever wondered what kind of data pathologists work or if you ever need to explain it, this is a talk for you. And if you have any questions, let me know in the comments below!

Another great show!

So early!!! Thank goodness for replay~!

Hi, Haruna Nggada from Maiduguri, Nigeria, Anatomic Pathologist

Dialled in from South Africa and always enjoy your sessions.

I m from India

Too bad I missed the live stream. I will try not to miss it again. Please rock on

Interested in doing AI straight from the microscope, get more info here: www.aleksandrazuraw.com/smart-in-media-9128

It was so much fun to be the guest this time :) Dennis Strenk, from "People of Pathology" was interviewing me about my book. Who is a listener of "People of Pathology" and heard the original episode?

Ive actualy told this to a doctor who has worked with the program like 7 years and he learned smth new that day. Last time i saw him he still used this method. This is also true for many commercial apps with a little preview like this, neat little design the developers incorperated to make life a little easier

Thank you Aleks for stopping at our booth. It was a pleasure meeting you in person. We continue polishing the platform to improve user/editors/contributors' experience. Very soon, we will start working on establishing infrastructure for Breast Tumors collections which is planned to be launched in 2025. Sincerely, Sasha

Thank you so much

I was waiting for microscopic findings

Thank you

The Abstracts reviewed today: 1)Molecular Classification of Breast Cancer Using Weakly Supervised Learning (pubmed.ncbi.nlm.nih.gov/38938010/) 2)Clinical implementation of artificial-intelligence-assisted detection of breast cancer metastases in sentinel lymph nodes: the CONFIDENT-B single-center, non-randomized clinical trial (pubmed.ncbi.nlm.nih.gov/38937624/) 3)[The environmental impact of digital technology and artificial intelligence, in the time of digital pathology] (pubmed.ncbi.nlm.nih.gov/38937204/) 4) Digital versus Manual Tracing in Cephalometric Analysis: A Systematic Review and Meta-Analysis (pubmed.ncbi.nlm.nih.gov/38929786/) 5) Undermind for literature research: 2ly.link/1yjS6

You can access this repository here: www.worldtumorregistry.org/home

Thank you for this review !

Before I comment on your video, for context, I produced static technical illustrations as well as animated presentations showing assembly and disassembly of complex equipment. I enjoyed your enthusiasm, appreciate your knowledge and the fact that you are making videos about obtaining microscopic images using our phones. Now here's my feedback and I hope it is useful: It's true that movement creates energy and is likely to draw someone into a video more than not doing that. But since you are teaching assembly and disassembly and showing different pieces of equipment, it might be useful to insert slides that use illustrations or higher contrast photos to show orientation and order of assembly. If you have multiple illustrations you could even use a different color for each part so that when you show the assembled result, It's easy to see what went where. If you can't use illustrations, just do these two things: 1. Speak more slowly 2. When you show us something, show it close up and for a bit longer. I have not checked your newer videos to see if you have already changed the way that you describe other activities. I hope that this information is useful for you and that you continue to provide such excellent content.

Here are the abstracts we reviewed today: 1. Prostate Cancer Risk Stratification by Digital Histopathology and Deep Learning, JCO Clin Cancer Inform. 2024 Jun:8:e2300184. doi: 10.1200/CCI.23.00184. 2. A pathologist-AI collaboration framework for enhancing diagnostic accuracies and efficiencies Nat Biomed Eng. 2024 Jun 19. doi: 10.1038/s41551-024-01223-5. Online ahead of print. 3. AI powered quantification of nuclear morphology in cancers enables prediction of genome instability and prognosis NPJ Precis Oncol. 2024 Jun 19;8(1):134. doi: 10.1038/s41698-024-00623-9. 4. Performance evaluation of a novel artificial intelligence (AI)-assisted digital microscopy system for the routine analysis of bone marrow aspirates Mod Pathol. 2024 Jun 17:100542. doi: 10.1016/j.modpat.2024.100542. Online ahead of print.

I graduated from Med school in 2009. The pathology department at the Medical University of South Carolina was great, and partially responsible for me pursuing a career in pathology. I loved learning path in med school.

This is a very good real life demonstration 😊

Good to see other people dressed in puffer-jackets indoors:)

What I'm most concerned about is the cost of this system?

plz comment on melanocyte.difficult to differentiate from keratinocyte

I saw a study of use of chatgpt in pathology on twitter where the bot was analysing histopath images. I was wowed. Maybe we can focus more in that regard and improve it

How are you using chatGPT? Me: literature research, draft writing, presentation outlines and basically anything I can think of :)

I cant understant to read inglish

I cubano how life in the island.i don't know abaut this linea I need very happy to ser and lisen You to speak abaut pathology. Thanks You.