SEARCH · PubMed Health

Results for “artificial intelligence (AI)”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 289 records · Page 16Linked to original sources

PubMed2002

Artificial life and the Chinese room argument.

"Strong artificial life" refers to the thesis that a sufficiently sophisticated computer simulation of a life form is a life form in its own right. Can John Searle's Chinese room argument [12]-originally intended by him to show that the thesis he dubs "strong AI" is false-be deployed against strong ALife? We have often encountered the suggestion that it can be (even in print; see Harnad [8]). We do our best to transfer the argument from the domain of AI to that of ALife. We do so in order to show once and for all that the Chinese room argument proves nothing about ALife. There may indeed be powerful philosophical objections to the thesis of strong ALife, but the Chinese room argument is not among them.

Artificial Intelligence↗

PubMed2003

Intelligence with representation.

Behaviour-based robotics has always been inspired by earlier cybernetics work such as that of W. Grey Walter. It emphasizes that intelligence can be achieved without the kinds of representations common in symbolic AI systems. The paper argues that such representations might indeed not be needed for many aspects of sensory-motor intelligence but become a crucial issue when bootstrapping to higher levels of cognition. It proposes a scenario in the form of evolutionary language games by which embodied agents develop situated grounded representations adapted to their needs and the conventions emerging in the population.

Adaptation, Physiological↗

PubMed2026

Comparative Efficacy of Different AI Systems for Polyp Detection by Size During Colonoscopy: Systematic Review and Network Meta-Analysis.

BACKGROUND: Colorectal cancer remains a leading cause of death despite being largely preventable through polypectomy. AI systems designed to enhance polyp detection during colonoscopy have shown promise, but the extent to which they improve detection of different-sized polyps remains unclear. OBJECTIVE: This study compared the size-stratified efficacy of AI-assisted colonoscopy vs standard colonoscopy using the Hartung-Knapp-Sidik-Jonkman (HKSJ) method, and generated exploratory rankings while acknowledging all cross-platform comparisons are indirect. METHODS: This systematic review and network meta-analysis (NMA) searched PubMed, Embase, Cochrane CENTRAL, and Web of Science from inception to July 25, 2026, supplemented by citation searching. We included randomized controlled trials (RCTs) comparing AI-assisted vs standard colonoscopy in adults (≥18 years of age), reporting mean polyp detection counts stratified by size (≤5 mm, 6-9 mm, and ≥10 mm). Two reviewers screened studies, extracted data, and assessed risk of bias using the Cochrane Risk of Bias 2.0. We conducted frequentist NMA using the HKSJ method with restricted maximum likelihood estimation, calculated 95% prediction intervals (PIs), and assessed heterogeneity using I2 and τ2. Certainty of evidence was rated using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework. RESULTS: A total of 13 RCTs (4156 participants) compared 8 AI systems to standard colonoscopy, forming a network without direct AI comparisons. For diminutive polyps (≤5 mm), AI showed a modest advantage (standardized mean difference [SMD] 0.21, 95% CI 0.07 to 0.35, 95% PI -1.12 to 1.54), but substantial heterogeneity (I2=86.6%) and wide PI crossing the null indicated high uncertainty. EndoScreener showed the most consistent evidence (SMD 0.36, 95% CI 0.18-0.54). For small and large polyps, effects were minimal (SMD 0.02, 95% CI -0.02 to 0.06, 95% PI -0.03 to 0.07; SMD 0.01, 95% CI 0.00-0.02, 95% PI -0.01 to 0.03). GRADE certainty was very low for diminutive polyps and low for small and large polyps. Sensitivity analysis excluding Tianjin YuJin did not materially change findings. CONCLUSIONS: AI may modestly enhance diminutive polyp detection, but effects on small and large polyps are minimal, with no platform superiority. Given very low to low certainty, findings are hypothesis-generating. This exploratory NMA provides size-stratified comparisons that can inform future head-to-head trial design. Unlike prior reviews aggregating all polyp sizes, we show the overall AI benefit is driven by diminutive polyp detection, providing a framework for targeted deployment-prioritizing AI for diminutive polyp screening, with limited value for larger lesions. Head-to-head trials are urgently needed. TRIAL REGISTRATION: PROSPERO International Prospective Register of Systematic Reviews CRD420251266932; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251266932.

Colonoscopy↗

PubMed2001

AIM: a personal view of where I have been and where we might be going.

My own career in medical informatics and AI in medicine has oscillated between concerns with medical records and concerns with knowledge representation with decision support as a pivotal integrating issue. It has focused on using AI to organise information and reduce 'muddle' and improve the user interfaces to produce 'useful and usable systems' to help doctors with a 'humanly impossible task'. Increasingly knowledge representation and ontologies have become the fulcrum for orchestrating re-use of information and integration of systems. Encouragingly, the dilemma between computational tractability and expressiveness is lessening, and ontologies and description logics are joining the mainstream both in AI in Medicine and in Intelligent Information Management generally. It has been shown possible to scale up ontologies to meet medical needs, and increasingly ontologies are playing a key role in meeting the requirements to scale up the complexity of clinical systems to meet the ever increasing demands brought about by new emphasis on reduction of errors, clinical accountability, and the explosion of knowledge on the Web.

Artificial Intelligence↗

PubMed2026

scBaseCount: An AI agent-curated, standardized, auto-updated single-cell data repository.

Single-cell RNA sequencing has transformed cell biology by enabling precise transcriptomic measurements of individual cells. The Sequence Read Archive (SRA) is the largest public repository of sequencing reads, yet much of it remains underutilized due to unstandardized metadata. Here, we introduce scBaseCount, a database that leverages an AI agent to automate discovery and metadata extraction and standardize data processing. Built by mining all 10x Genomics datasets, scBaseCount is the largest public repository of single-cell gene expression data, comprising over 502 million cells across 27 organisms and 75 tissues. It offers an unbiased view of the data landscape within the SRA and enables the training of more performant computational models through access to broader phenotypic diversity. Uniform processing enables measurement of both intronic and exonic reads and non-coding gene expression and improves alignment across experiments. Moreover, scBaseCount provides a blueprint for how AI can be leveraged to autonomously curate biological data repositories.

Single-Cell Analysis↗

PubMed1998

Assessing an AI knowledge-base for asymptomatic liver diseases.

Discovering not yet seen knowledge from clinical data is of importance in the field of asymptomatic liver diseases. Avoidance of liver biopsy which is used as the ultimate confirmation of diagnosis by making the decision based on relevant laboratory findings only, would be considered an essential support. The system based on Quinlan's ID3 algorithm was simple and efficient in extracting the sought knowledge. Basic principles of applying the AI systems are therefore described and complemented with medical evaluation. Some of the diagnostic rules were found to be useful as decision algorithms i.e. they could be directly applied in clinical work and made a part of the knowledge-base of the Liver Guide, an automated decision support system.

Algorithms↗

PubMed2004

Filtering of ineffective siRNAs and improved siRNA design tool.

MOTIVATION: Short interfering RNAs (siRNAs) can be used to suppress gene expression and possess many potential applications in therapy, but how to design an effective siRNA is still not clear. Based on the MPI (Max-Planck-Institute) basic principles, a number of siRNA design tools have been developed recently. The set of candidates reported by these tools is usually large and often contains ineffective siRNAs. In view of this, we initiate the study of filtering ineffective siRNAs. RESULTS: The contribution of this paper is 2-fold. First, we propose a fair scheme to compare existing design tools based on real data in the literature. Second, we attempt to improve the MPI principles and existing tools by an algorithm that can filter ineffective siRNAs. The algorithm is based on some new observations on the secondary structure, which we have verified by AI techniques (decision trees and support vector machines). We have tested our algorithm together with the MPI principles and the existing tools. The results show that our filtering algorithm is effective. AVAILABILITY: The siRNA design software tool can be found in the website http://www.cs.hku.hk/~sirna/ CONTACT: smyiu@cs.hku.hk

Algorithms↗

PubMed1995

Consciousness: perspectives from symbolic and connectionist AI.

While consciousness has not been a major concern of most AI researchers, some theorists have tried to explore how computational models might explain consciousness. I explore how far computational models might go in explaining consciousness, focusing on three aspects of conscious mental states: their intrinsic intentionality, the awareness a subject has of the contents of these intentional states, and the distinctive qualitative character of these states. I describe and evaluate strategies for developing connectionist systems that satisfy these aspects of consciousness.

Animals↗

PubMed2026

Effectiveness of an AI-based home exercise app for rehabilitation of rotator cuff-related shoulder pain: A randomized controlled trial.

BACKGROUND: Rotator cuff-related shoulder pain contributes to disability and healthcare use. Although therapeutic exercise is first-line treatment, limited supervision and adherence may reduce its effectiveness; digital rehabilitation with real-time feedback may address these limitations. OBJECTIVES: To evaluate the effectiveness of adding a digital rehabilitation program to standard physiotherapy on pain, function, fear-avoidance beliefs, and healthcare utilization. DESIGN: Single-center, assessor-blinded, randomized controlled trial with two parallel groups. METHOD: Forty-six adults (mean age 59 years) with rotator cuff-related shoulder pain were randomized to 12 weeks of conventional physiotherapy or physiotherapy plus an AI-based digital rehabilitation program using computer vision for real-time feedback and performance monitoring. Outcomes were assessed at baseline and at 2, 4, and 12 weeks. Pain intensity (NPRS) was primary outcome; secondary outcomes included upper limb function (QuickDASH), fear-avoidance beliefs (FABQ), and post-intervention healthcare utilization. Analyses followed an intention-to-treat approach. RESULTS: Pain reduction exceeded the MCID (1.3) at 4 and 12 weeks. Between-group differences favoured the intervention at Weeks 2 and 4 (MD -0.7; 95% CI -1.13 to -0.14 and MD -1.01; 95% CI -1.8 to -0.2, respectively). Upper limb function improved more at Week 4 (MD -7.3; 95% CI -12.3 to -2.2). FABQ scores decreased more at Week 12 (MD -7.6; 95% CI -14 to -0.5). Fewer participants in the experimental group required post-intervention healthcare (3 vs 10; p = 0.02). CONCLUSION: Adding AI-based home exercise app to conventional treatment improve pain and may improve function and reduce healthcare utilization in rotator cuff-related shoulder pain.

Humans↗

PubMed2006

Rule generation for protein secondary structure prediction with support vector machines and decision tree.

Support vector machines (SVMs) have shown strong generalization ability in a number of application areas, including protein structure prediction. However, the poor comprehensibility hinders the success of the SVM for protein structure prediction. The explanation of how a decision made is important for accepting the machine learning technology, especially for applications such as bioinformatics. The reasonable interpretation is not only useful to guide the "wet experiments," but also the extracted rules are helpful to integrate computational intelligence with symbolic AI systems for advanced deduction. On the other hand, a decision tree has good comprehensibility. In this paper, a novel approach to rule generation for protein secondary structure prediction by integrating merits of both the SVM and decision tree is presented. This approach combines the SVM with decision tree into a new algorithm called SVM_ DT, which proceeds in three steps. This algorithm first trains an SVM. Then, a new training set is generated through careful selection from the output of the SVM. Finally, the obtained training set is used to train a decision tree learning system and to extract the corresponding rule sets. The results of the experiments of protein secondary structure prediction on RS126 data set show that the comprehensibility of SVM_DT is much better than that of the SVM. Moreover, the generalization ability of SVM_DT is better than that of C4.5 decision trees and is similar to that of the SVM. Hence, SVM_DT can be used not only for prediction, but also for guiding biological experiments.

Algorithms↗

PubMed2026

Identification of compounds that repress DUX4 expression in facioscapulohumeral muscular dystrophy.

Facioscapulohumeral muscular dystrophy (FSHD) is caused by epigenetic dysregulation of the disease locus, leading to pathogenic misexpression of DUX4 in skeletal muscle. Thus, most FSHD therapeutic approaches target DUX4. Our previous study identified the chromatin remodeling factor BAZ1A (bromodomain adjacent to zinc finger domain protein 1A) as a promising target for therapeutic development. Here we used an artificial intelligence-based screening pipeline to identify molecules predicted to bind the BAZ1A bromodomain, and validated hit compounds using FSHD-specific assays in FSHD myocytes. One compound, termed C06, emerged as a potent repressor of DUX4 and DUX4 target gene expression. Interestingly, while C06 exhibited binding to BAZ1A in vitro, it can also inhibit multiple kinases, including p38α, an upstream activator of DUX4. Despite this, at low doses C06 was an equally effective and more specific repressor of DUX4 than losmapimod, which is a robust and specific p38 inhibitor. At low concentrations, C06 returns the DUX4 gene expression signature to a healthier profile without major effects on the muscle transcriptome. Thus, C06 is a useful tool for potent and specific DUX4 suppression, and a viable candidate for further development. Our results highlight both the utility and limitations of AI for targeted drug discovery, and the importance of using an FSHD-specific functional screening strategy for selecting relevant candidates.

Muscular Dystrophy, Facioscapulohumeral↗

PubMed1999

Is imitation learning the route to humanoid robots?

This review investigates two recent developments in artificial intelligence and neural computation: learning from imitation and the development of humanoid robots. It is postulated that the study of imitation learning offers a promising route to gain new insights into mechanisms of perceptual motor control that could ultimately lead to the creation of autonomous humanoid robots. Imitation learning focuses on three important issues: efficient motor learning, the connection between action and perception, and modular motor control in the form of movement primitives. It is reviewed here how research on representations of, and functional connections between, action and perception have contributed to our understanding of motor acts of other beings. The recent discovery that some areas in the primate brain are active during both movement perception and execution has provided a hypothetical neural basis of imitation. Computational approaches to imitation learning are also described, initially from the perspective of traditional AI and robotics, but also from the perspective of neural network models and statistical-learning research. Parallels and differences between biological and computational approaches to imitation are highlighted and an overview of current projects that actually employ imitation learning for humanoid robots is given.

Journal Article↗

PubMed1993

Building large knowledge bases in molecular biology.

Large scale genome sequencing projects are now producing hugh amounts of data which can be readily stored and managed within data base management systems, and analyzed using dedicated software packages. The results of these analyzes should also be stored with the input DNA sequences. The increasing complexity and size of the objects to be described and managed have led biologists to rely on advanced data models such as the object-oriented model. As a joint effort between our computer science and molecular biology research projects, the knowledge bases we have developed in molecular genetics have shown however that the basic object-oriented model is not fully adapted to the complexity of some biological situations encountered. Advanced descriptive capabilities, provided only by knowledge models originated from the AI field, are required. Composite or evolving objects, multiple viewpoints, constraints, tasks and methods, textual annotations are some examples of such capabilities. They are illustrated by biological situations for which they appeared to be necessary. Supporting powerful reasoning mechanisms (e.g. object classification, constraint propagation or qualitative simulators), they allow the development of large knowledge bases in molecular biology. These knowledge bases are expected to become the adequate support for co-operative distributed research efforts.

Artificial Intelligence↗

PubMed2026

An AI-assisted Clinical Decision Support System for Green Classification of Cystocele on Dynamic Transperineal Ultrasound.

Green classification of cystocele on dynamic transperineal ultrasound (TPUS) remains operator-dependent because it requires manual frame selection and landmark-based assessment of the Valsalva maneuver. We developed a workflow-oriented AI-assisted clinical decision support system for automated urethrovesical junction localization and dynamic Green classification and prospectively evaluated its standalone and reader-support performance. This diagnostic accuracy and reader study included 881 patients from a tertiary referral hospital, comprising a retrospective development cohort (n = 688) and an independent prospective test cohort (n = 193). A nested subset of 67 prospective patients was used for a reader study involving two junior and two intermediate radiologists under unaided and AI-assisted conditions. In the complete prospective test cohort, Green-AttGRU achieved a macro-averaged AUC of 0.939 (95% CI, 0.897-0.971) and an overall accuracy of 0.902 (95% CI, 0.860-0.943). In the reader study, overall accuracy increased from 0.761 to 0.821 without AI to 0.851-0.881 with AI, while macro-F1 increased from 0.660 to 0.777 to 0.820-0.860. Overall inter-reader agreement increased from a Fleiss' κ of 0.453 to 0.786, and pooled median interpretation time decreased from 26.7 s to 9.9 s. These findings support the preliminary feasibility of the system as a workflow-oriented decision-support tool for dynamic TPUS interpretation.

Humans↗

PubMed2005

An ontological approach to modelling tasks and goals.

Current approaches to modelling plans and processes in AI are limited by current understanding of "goals" and "intentions". We discuss this question in a medical context, viewing goals as clinical objectives and plans and processes as collections of tasks to achieve those objectives. The specific context for this discussion is the CREDO project, which aims to develop a comprehensive approach to supporting complex treatment plans and care pathways and provides an opportunity to carry out an extensive investigation of the key problem of formalising goals in medical informatics and AI generally. A review of 222 clinical services required in the management of breast cancer has yielded a systematic classification of tasks involved in breast cancer care and a similar classification of the clinical goals associated with these tasks. A six-component model of goal structure is proposed based on these ontological analyses. Although the model has been derived from a corpus of examples from medicine many of the issues encountered, and the solution proposed, appear to be relevant to other domains.

Artificial Intelligence↗

PubMed2025

Development and validation of a deep learning model based on cascade mask regional convolutional neural network to noninvasively and accurately identify human round spermatids.

INTRODUCTION: The difficulty of identifying human round spermatids (hRSs) has impeded applications of the human round spermatid injection (ROSI) technique. RSs can be accurately screened through flow cytometric analysis utilizing the Hoechst fluorescence profile reflecting DNA, but this method is not suitable for isolating hRSs due to the toxicity associated with Hoechst staining. OBJECTIVE: To evaluate the capacity of a deep learning model grounded in a cascade mask region-based convolutional neural network (R-CNN) for the noninvasive and accurate identification of hRSs. METHODS: In this study, we presented the development and validation of a deep learning model for identifying hRSs through the analysis of 3457 optical light microscope images of sorted hRSs obtained via flow cytometric analysis. The model's accuracy and specificity were evaluated by calculating the mean average precision (mAP). Furthermore, a double-blind experiment was conducted to access the reliability of the proposed model in accurately identifying hRSs. It detected the expression of protamine (PRM1) and/or peanut lectin (PNA), which are established markers for RSs. RESULTS: Our deep learning-based model demonstrated a high precision, achieving a mAP of over 0.80 for isolating hRSs in test datasets. The expression of PRM1 and/or PNA was observed in all cells noninvasively selected by our AI model during an independent double-blind test. This phenomenon confirmed the accuracy and effectiveness of the proposed model. The model's capability for noninvasive and accurate isolation of hRSs among spermatogenic cells highlighted its robustness and generalizability for clinical applications. CONCLUSION: The deep learning AI model based on a cascade R-CNN has the ability to accurately identify hRSs among spermatogenic cells. The application of this noninvasive method, which requires no additional procedures in clinical practice, is able to facilitate the widespread implementation of ROSI technique. Therefore, it can provide patients with spermatogenic arrest the opportunity to become biological fathers.

Humans↗

PubMedDate not supplied

Disseminating medical knowledge: the PROforma approach.

Medical knowledge is traditionally disseminated via the publication of documents and through participation in clinical practice. Information technology offers to extend both modes of dissemination, via electronic publishing and virtual reality training, for example. AI promises even more radical changes through the possibility of publishing clinical expertise in the form of expert systems, which assist patient care through active decision support and workflow management. PROforma is a knowledge representation language that is designed to support this new mode of dissemination. It is based on an intuitive model of the processes of care and well-understood logical semantics. This paper provides a description of the language and associated software tools, and discusses its potential roles in, and implications for, medical knowledge publishing.

Artificial Intelligence↗

PubMed1998

An AI approach to dynamic visual field testing.

Visual field test results are crucial to the accuracy and efficiency of diagnosing blinding diseases such as glaucoma. Herein, a method of integrating self-organizing neural networks and empirical heuristics is used to perform visual field tests via a dynamic test strategy, which can lead to a reduction in the number of trials in a perimetric test. Experiments performed using clinical test records show that we are able to reduce by 20% to 30% the number of trials per test without much adverse effect on the accuracy of the tests.

Algorithms↗