The tumour histopathology “glossary” for AI developers

PLoS Computational Biology, Journal Year: 2025, Volume and Issue: 21(1), P. e1012708 - e1012708

Published: Jan. 23, 2025

The applications of artificial intelligence (AI) and deep learning (DL) are leading to significant advances in cancer research, particularly analysing histopathology images for prognostic treatment-predictive insights. However, effective translation these computational methods requires researchers have at least a basic understanding histopathology. In this work, we aim bridge that gap by introducing essential concepts support AI developers their research. We cover the defining features key cell types, including epithelial, stromal, immune cells. malignancy, precursor lesions, tumour microenvironment (TME) discussed illustrated. To enhance understanding, also introduce foundational techniques, such as conventional staining with hematoxylin eosin (HE), antibody immunohistochemistry, new multiplexed methods. By providing knowledge community, accelerate development algorithms

Language: Английский

---

Artificial intelligence for diagnosis and predictive biomarkers in Non-Small cell lung cancer Patients: New promises but also new hurdles for the pathologist

Lung Cancer, Journal Year: 2025, Volume and Issue: 200, P. 108110 - 108110

Published: Jan. 27, 2025

Language: Английский

---

The foundational capabilities of large language models in predicting postoperative risks using clinical notes

npj Digital Medicine, Journal Year: 2025, Volume and Issue: 8(1)

Published: Feb. 11, 2025

Clinical notes recorded during a patient's perioperative journey holds immense informational value. Advances in large language models (LLMs) offer opportunities for bridging this gap. Using 84,875 preoperative and its associated surgical cases from 2018 to 2021, we examine the performance of LLMs predicting six postoperative risks using various fine-tuning strategies. Pretrained outperformed traditional word embeddings by an absolute AUROC 38.3% AUPRC 33.2%. Self-supervised further improved 3.2% 1.5%. Incorporating labels into training increased 1.8% 2%. The highest was achieved with unified foundation model, improvements 3.6% 2.6% compared self-supervision, highlighting foundational capabilities risks, which could be potentially beneficial when deployed care.

Language: Английский

---

Artificial intelligence-assisted wearable electronics for human-machine interfaces

Device, Journal Year: 2025, Volume and Issue: unknown, P. 100707 - 100707

Published: Feb. 1, 2025

Language: Английский

---

Artificial Intelligence-Based Algorithm for Pathological Response Assessment to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer: A Multi-Center Derived Platform Can Shrink a Mountainous Task into a Molehill

Published: Jan. 1, 2025

Language: Английский

---

Evaluating interactions of patients with large language models for medical information

BJU International, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 18, 2025

To explore the interaction of real-world patients with a chatbot in clinical setting, investigating key aspects medical information provided by large language models (LLMs). The study enrolled 300 seeking urological counselling between February and July 2024. First, participants voluntarily conversed Generative Pre-trained Transformer 4 (GPT-4) powered to ask questions related their situation. In following survey, rated perceived utility, completeness, understandability during simulated conversation as well user-friendliness. Finally, were asked which, experience, best answered questions: LLMs, urologists, or search engines. A total 292 completed study. majority providing useful, complete, understandable information, being user-friendly. However, ability human urologists answer an way was higher than LLMs. Interestingly, 53% question-answering LLMs Age not associated preferences. Limitations include social desirability sampling biases. This highlights potential enhance patient education communication settings, valuing user-friendliness comprehensiveness for information. By addressing preliminary questions, could potentially relieve time constraints on healthcare providers, enabling personnel focus complex inquiries care.

Language: Английский

---

Role of circular RNAs in cancer therapy resistance

Molecular Cancer, Journal Year: 2025, Volume and Issue: 24(1)

Published: Feb. 25, 2025

Over the past decade, circular RNAs (circRNAs) have gained recognition as a novel class of genetic molecules, many which are implicated in cancer pathogenesis via different mechanisms, including drug resistance, immune escape, and radio-resistance. ExosomalcircRNAs, particular, facilitatecommunication between tumour cells micro-environmental cells, fibroblasts, other components. Notably, can reportedly influence progression treatment resistance by releasing exosomalcircRNAs. circRNAs often exhibit tissue- cancer-specific expression patterns, growing evidence highlights their potential clinical relevance utility. These molecules show strong promise biomarkers therapeutic targets for diagnosis treatment. Therefore, this review aimed to briefly discuss latest findings on roles mechanisms key various malignancies, lung, breast, liver, colorectal, gastric cancers, well haematological malignancies neuroblastoma.This will contribute identification new circRNA early cancer.

Language: Английский

---

Tumor-Agnostic Therapies in Practice: Challenges, Innovations, and Future Perspectives

Cancers, Journal Year: 2025, Volume and Issue: 17(5), P. 801 - 801

Published: Feb. 26, 2025

This review comprehensively analyzes the current landscape of tumor-agnostic therapies in oncology. Tumor-agnostic are designed to target specific molecular alterations rather than primary site tumor, representing a shift cancer treatment. We discuss recent approvals by regulatory agencies such as FDA and EMA, highlighting that have demonstrated efficacy across multiple types sharing common alterations. delve into trial methodologies underpin these approvals, emphasizing innovative designs basket trials umbrella trials. These present unique advantages, including increased efficiency patient recruitment ability assess drug diverse populations rapidly. However, they also entail certain challenges, need for robust biomarkers complexities requirements. Moreover, we examine promising prospects developing rare cancers exhibit targets typically associated with more prevalent malignancies. By synthesizing insights, this underscores transformative potential It offers pathway personalized treatment transcends conventional histology-based classification.

Language: Английский

---

Practical Applications of Artificial Intelligence Diagnostic Systems in Fundus Retinal Disease Screening

International Journal of General Medicine, Journal Year: 2025, Volume and Issue: Volume 18, P. 1173 - 1180

Published: Feb. 28, 2025

This study aims to evaluate the performance of a deep learning-based artificial intelligence (AI) diagnostic system in analysis retinal diseases, assessing its consistency with expert diagnoses and overall utility screening applications. A total 3076 patients attending our hospital underwent comprehensive ophthalmic examinations. Initial assessments were performed using AI, Comprehensive AI Retinal Expert (CARE) system, followed by thorough manual reviews establish final diagnoses. comparative was conducted between AI-generated results evaluations senior ophthalmologists assess reliability feasibility context screening. : The demonstrated sensitivity 94.12% specificity 98.60% for diabetic retinopathy (DR); 89.50% 98.33% age-related macular degeneration (AMD); 91.55% 97.40% suspected glaucoma; 90.77% 99.10% pathological myopia; 81.58% 99.49% vein occlusion (RVO); 88.64% 99.18% detachment; 83.33% 99.80% hole; 82.26% 99.23% epiretinal membrane; 94.55% 97.82% hypertensive retinopathy; 99.74% myelinated fibers; 75.00% 99.95% retinitis pigmentosa. Additionally, exhibited notable other prevalent conditions, including DR, glaucoma, myopia, retinopathy. AI-assisted exhibits high majority suggesting potential as valuable tool practices. Its implementation is particularly beneficial grassroots community healthcare settings, facilitating initial efforts enhancing efficacy tiered care, important implications broader clinical adoption.

Language: Английский

---

Knowledge-guided diffusion model for 3D ligand-pharmacophore mapping

Nature Communications, Journal Year: 2025, Volume and Issue: 16(1)

Published: March 6, 2025

Pharmacophores are abstractions of essential chemical interaction patterns, holding an irreplaceable position in drug discovery. Despite the availability many pharmacophore tools, adoption deep learning for pharmacophore-guided discovery remains relatively rare. We herein propose a knowledge-guided diffusion framework 'on-the-fly' 3D ligand-pharmacophore mapping, named DiffPhore. It leverages matching knowledge to guide ligand conformation generation, meanwhile utilizing calibrated sampling mitigate exposure bias iterative search process. By training on two self-established datasets pairs, DiffPhore achieves state-of-the-art performance predicting binding conformations, surpassing traditional tools and several advanced docking methods. also manifests superior virtual screening power lead target fishing. Using DiffPhore, we successfully identify structurally distinct inhibitors human glutaminyl cyclases, their modes further validated through co-crystallographic analysis. believe this work will advance AI-enabled techniques.

Language: Английский

---