Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “artificial intelligence”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 1,063 records · Page 59Linked to original sources
PubMed2002

Automated planning volume definition in soft-tissue sarcoma adjuvant brachytherapy.

In current practice, the planning volume for adjuvant brachytherapy treatment for soft-tissue sarcoma is either not determined a priori (in this case, seed locations are selected based on isodose curves conforming to a visual estimate of the planning volume), or it is derived via a tedious manual process. In either case, the process is subjective and time consuming, and is highly dependent on the human planner. The focus of the work described herein involves the development of an automated contouring algorithm to outline the planning volume. Such an automatic procedure will save time and provide a consistent and objective method for determining planning volumes. In addition, a definitive representation of the planning volume will allow for sophisticated brachytherapy treatment planning approaches to be applied when designing treatment plans, so as to maximize local tumour control and minimize normal tissue complications. An automated tumour volume contouring algorithm is developed utilizing computational geometry and numerical interpolation techniques in conjunction with an artificial intelligence method. The target volume is defined to be the slab of tissue r cm perpendicularly away from the curvilinear plane defined by the mesh of catheters. We assume that if adjacent catheters are over 2r cm apart, the tissue between the two catheters is part of the tumour bed. Input data consist of the digitized coordinates of the catheter positions in each of several cross-sectional slices of the tumour bed, and the estimated distance r from the catheters to the tumour surface. Mathematically, one can view the planning volume as the volume enclosed within a minimal smoothly-connected surface which contains a set of circles, each circle centred at a given catheter position in a given cross-sectional slice. The algorithm performs local interpolation on consecutive triplets of circles. The effectiveness of the algorithm is evaluated based on its performance on a collection of soft-tissue sarcoma tumour beds within various anatomical structures. For each of 15 patient cases considered, the algorithm takes approximately 2 min to generate the planning volume. Although the tumour shapes are rather different, the algorithm consistently generates planning volumes that visually demonstrate smooth curves compactly encapsulating the circles. This general-purpose contouring algorithm works well whether the catheters are all close together, spread far apart in the plane or arranged in a convoluted way. The automatic contouring algorithm significantly reduces labour time and provides a consistent and objective method for determining planning volumes for soft-tissue sarcoma. Further studies are needed to validate the significance of the resulting planning volumes in designing treatment plans and the role that sophisticated brachytherapy treatment planning optimization may have in producing good plans.

Adult↗
PubMed1989

Real time processing and analysis of fetal phonocardiographic signals.

The monitoring of fetal heart rate (FHR) is commonly used in assessing the general health of the fetus. Although certain periodic cycles may be indicative of fetal problems, only short term observations are routinely employed in clinical practice. This is due to cost considerations, inconvenience to the patient and concern about long term ultrasonic monitoring. Therefore only a low confidence assessment can be established between detected rhythms and the health of the fetus. The technique advocated in this paper makes use of an inexpensive, non-invasive phonocardiographic (phono) transducer which facilitates safe long-term patient monitoring. A variable comb filter applied to the frequency domain is used in order to take full advantage of the harmonic content of fetal heart signals. Real time estimation of FHR has been achieved on pre-recorded phono signals lasting eight hours. Recordings with a reasonable signal quality were analysed and some of the results are given. Advanced signal processing techniques followed by Artificial Intelligence (AI) algorithms reduce the number of erroneous estimates during periods of low signal to noise ration (SNR). The resulting FHR time series is stored on the host computer for further processing, display and parameter extraction. This paper outlines the processing steps involved.

Computer Systems↗
PubMed2001

Preliminary results from an EIT breast imaging simulation system.

An electrical impedance tomography (EIT) simulation system for breast imaging has been developed using impedance data from a previous study and a finite-element model (FEM). This system has the functionality to construct various models of the breast, image the boundary voltages developed from any injection schema and provide parametric Cole-Cole modelling. Simulations indicate that breast carcinoma can be imaged and multi-frequency Cole-Cole dispersion data can be extracted. This is intended as the first stage towards providing an artificial intelligence based system capable of producing clinically relevant analytical data.

Algorithms↗
PubMed2001

Use of artificial neural networks in prostate cancer.

Artificial neural networks (ANNs) are a type of artificial intelligence software inspired by biological neuronal systems that can be used for nonlinear statistical modeling. In recent years, these applications have played an increasing role in predictive and classification modeling in medical research. We review the basic concepts behind ANNs and examine the role of this technology in selected applications in prostate cancer research.

Humans↗
PubMed2006

A cognitive approach to game usability and design: mental model development in novice real-time strategy gamers.

We developed a technique to observe and characterize a novice real-time-strategy (RTS) player's mental model as it shifts with experience. We then tested this technique using an off-the-shelf RTS game, EA Games Generals. Norman defined mental models as, "an internal representation of a target system that provides predictive and explanatory power to the operator." In the case of RTS games, the operator is the player and the target system is expressed by the relationships within the game. We studied five novice participants in laboratory-controlled conditions playing a RTS game. They played Command and Conquer Generals for 2 h per day over the course of 5 days. A mental model analysis was generated using player dissimilarity-ratings of the game's artificial intelligence (AI) agents analyzed using multidimensional scaling (MDS) statistical methods. We hypothesized that novices would begin with an impoverished model based on the visible physical characteristics of the game system. As they gained experience and insight, their mental models would shift and accommodate the functional characteristics of the AI agents. We found that all five of the novice participants began with the predicted physical-based mental model. However, while their models did qualitatively shift with experience, they did not necessarily change to the predicted functional-based model. This research presents an opportunity for the design of games that are guided by shifts in a player's mental model as opposed to the typical progression through successive performance levels.

Cognition↗
PubMed1998

Recognition of splice junctions on DNA sequences by BRAIN learning algorithm.

MOTIVATION: The problem addressed in this paper is the prediction of splice site locations in human DNA. The aims of the proposed approach are explicit splicing rule description, high recognition quality, and robust and stable 'one shot' data processing. RESULTS: These results are achieved by means of a new learning algorithm [BRAIN (Batch Relevance-based Artificial INtelligence)], described in the paper, inferring Boolean formulae from examples, and by considering the splicing rules as disjunctive normal form (DNF) formulae. The formula terms are computed in an iterative way, by identifying from the training set a relevance coefficient for each attribute. The classification is then refined by a neural network and combined with a discriminant analysis procedure. This splice site recognition method shows low error rates (0.0002 and 0.0003) and high correlation coefficient measures (0.83 and 0.81) for donor and acceptor sites, respectively; better than other methods. AVAILABILITY: The BRAIN package (Borland Turbo Pascal for Windows) is available on the EMBL file server. (ftp://ftp.ebi.ac.uk/pub/software/dos under nnbrain$.exe). CONTACT: rampo@vaxsa.csied.unisa.it

Algorithms↗
PubMed1988

Algorithms for determining the fate of sites and domain boundaries in computer simulations of recombinant DNA procedures.

Structural and functional features within genomic sequences are best described by their position within the genomic structure. Cleavage sites can be conveniently described by single positions but genomic domains require the position of their two boundaries. The handling of these positions simultaneously to sequence manipulations in computer simulations of recombinant DNA procedures greatly improves the understanding of the resulting recombinant constructs. In addition, the algorithms describing the fate of domain boundaries can be used for the handling of nucleotide sequences in dynamic database environments handled by languages like Prolog which are particularly suitable for artificial intelligence implementations. This communication describes a set of algorithms for the automatic updating of single sites and double domain boundaries in linear and circular models for computer simulation of recombinant DNA procedures.

Algorithms↗
PubMed2025

SetBERT: the deep learning platform for contextualized embeddings and explainable predictions from high-throughput sequencing.

MOTIVATION: High-throughput sequencing (HTS) is a modern sequencing technology used to profile microbiomes by sequencing thousands of short genomic fragments from the microorganisms within a given sample. This technology presents a unique opportunity for artificial intelligence to comprehend the underlying functional relationships of microbial communities. However, due to the unstructured nature of HTS data, nearly all computational models are limited to processing DNA sequences individually. This limitation causes them to miss out on key interactions between microorganisms, significantly hindering our understanding of how these interactions influence the microbial communities as a whole. Furthermore, most computational methods rely on post-processing of samples which could inadvertently introduce unintentional protocol-specific bias. RESULTS: Addressing these concerns, we present SetBERT, a robust pre-training methodology for creating generalized deep learning models for processing HTS data to produce contextualized embeddings and be fine-tuned for downstream tasks with explainable predictions. By leveraging sequence interactions, we show that SetBERT significantly outperforms other models in taxonomic classification with genus-level classification accuracy of 95%. Furthermore, we demonstrate that SetBERT is able to accurately explain its predictions autonomously by confirming the biological-relevance of taxa identified by the model. AVAILABILITY AND IMPLEMENTATION: All source code is available at https://github.com/DLii-Research/setbert. SetBERT may be used through the q2-deepdna QIIME 2 plugin whose source code is available at https://github.com/DLii-Research/q2-deepdna.

Deep Learning↗
PubMed2025

Study research protocol for Phenome India-CSIR Health Cohort Knowledgebase: A prospective multi-modal follow-up study on a nationwide employee cohort.

Predicting individual health trajectories based on risk scores can help formulate effective preventive strategies for diseases and their complications. Currently, most risk prediction algorithms rely on epidemiological data from the Caucasian population, which often do not translate well to the Indian population due to ethnic diversity, differing dietary and lifestyle habits, and unique risk profiles. In this multi-center prospective longitudinal study conducted across India, we aim to address these challenges by developing clinically relevant risk prediction scores for cardio-metabolic diseases specifically tailored to the Indian population. India, which accounts for nearly 18% of the global population, also has a significant diaspora worldwide. This program targets longitudinal collection and bio-banking of samples from over 10 000 employees both working and retirees of the Council of Scientific and Industrial Research and their spouses, with baseline sample collection already completed. During the baseline collection, we gathered multi-parametric data including clinical questionnaires, lifestyle and dietary habits, anthropometric parameters, lung function assessments, liver elastography by Fibroscan, electrocardiogram readings, biochemical data, and molecular assays, including but not limited to genomics, plasma proteomics, metabolomics, and fecal microbiome analysis. In addition to exploring associations between these parameters and their cardio-metabolic outcomes, we plan to employ artificial intelligence algorithms to develop predictive models for phenotypic conditions. This study could pave the way for precision medicine tailored to the Indian population, particularly for the middle-income strata, and help refine the normative values for health and disease indicators in India.

cardio-metabolic↗
PubMed2001

Consciousness.

Consciousness is topical, for reasons including its renewed respectability among psychologists, rapid progress in the neuroscience of perception, memory and action, advances in artificial intelligence and dissatisfaction with the dualistic separation of mind and body. Consciousness is an ambiguous term. It can refer to (i) the waking state; (ii) experience; and (iii) the possession of any mental state. Self-consciousness is equally ambiguous, with senses including (i) proneness to embarrassment in social settings; (ii) the ability to detect our own sensations and recall our recent actions; (iii) self-recognition; (iv) the awareness of awareness; and (v) self-knowledge in the broadest sense. The understanding of states of consciousness has been transformed by the delineation of their electrical correlates, of structures in brainstem and diencephalon which regulate the sleep-wake cycle, and of these structures' cellular physiology and regional pharmacology. Clinical studies have defined pathologies of wakefulness: coma, the persistent vegetative state, the 'locked-in' syndrome, akinetic mutism and brain death. Interest in the neural basis of perceptual awareness has focused on vision. Increasingly detailed neuronal correlates of real and illusory visual experience are being defined. Experiments exploiting circumstances in which visual experience changes while external stimulation is held constant are tightening the experimental link between consciousness and its neural correlates. Work on unconscious neural processes provides a complementary approach. 'Unperceived' stimuli have detectable effects on neural events and subsequent action in a range of circumstances: blindsight provides the classical example. Other areas of cognitive neuroscience also promise experimental insights into consciousness, in particular the distinctions between implicit and explicit memory and deliberate and automatic action. Overarching scientific theories of consciousness include neurobiological accounts which specify anatomical or physiological mechanisms for awareness, theories focusing on the role played by conscious processes in information processing and theories envisaging the functions of consciousness in a social context. Whether scientific observation and theory will yield a complete account of consciousness remains a live issue. Physicalism, functionalism, property dualism and dual aspect theories attempt to do justice to three central, but controversial, intuitions about experience: that it is a robust phenomenon which calls for explanation, that it is intimately related to the activity of the brain and that it has an important influence on behaviour.

Animals↗
PubMed2026

Conference report: the third Bacterial Genome Sequencing Pan-European Network conference.

The third Bacterial Genome Sequencing Pan-European Network conference, held in Engelberg, Switzerland (12-15 January 2026), brought together experts from six European countries to discuss the implementation of bacterial genome sequencing in clinical microbiology and public health. Key themes included regulatory frameworks (In Vitro Diagnostic Regulation, General Data Protection Regulation), standardization, quality control, data sharing, economic evaluation, and the integration of artificial intelligence and long-read sequencing into diagnostic workflows. Across presentations, panel discussions, and workshops, participants emphasized that successful implementation of genome sequencing requires more than technical capacity: it depends on robust validation, sustainable funding, interoperable data standards, ethical governance, and interdisciplinary collaboration. The meeting highlighted that sequencing should remain question-driven and clinically meaningful, balancing cost, turnaround time, and public health impact. Overall, the conference reinforced the need for coordinated European efforts to advance responsible, standardized, and sustainable genomic surveillance and diagnostics.

bacterial genome sequencing↗
PubMed2026

Advances in genomics-driven genetic decoding and genomic design breeding in tomato.

Tomatoes are highly nutritious and represent one of the important vegetable fruits worldwide. Both historically and moving forward, genetic decoding and precision breeding remain fundamental to tomato improvement. Here, we summarize pivotal advances in decoding tomato genomes across domestication, improvement and evolution processes and provide a perspective on future breeding through precision design. In-depth population genetic studies have revealed how artificial selection systematically prioritized yield-related alleles at the cost of narrowing genetic diversity, especially at flavor-related loci-highlighting the urgent need to reconcile these trade-offs. Comparative genomics across species, viewed through an evolutionary lens, has uncovered critical insights into functional genes, deepening our understanding of the genetic architecture and regulatory mechanisms underlying key traits. Collectively, these advances have enabled precise identification and functional characterization of key genetic elements, paving the way for systematic redomestication of tomato through precision genomic design. Looking ahead, more efficient and precise breeding strategies will be required to accelerate genetic gains in tomato in the coming decades. The integration of recent genomic advances, coupled with genomic selection and artificial intelligence, into genomic design breeding offers a transformative framework, unlocking unprecedented opportunities for developing highly flavorful and consumer-customized tomato varieties.

Journal Article↗
PubMed2026

Mapping ovarian cellular and molecular landscape across the lifespan of women: a scoping review.

BACKGROUND: With growing interest in ART, fertility preservation, and postmenopausal health of women, reproductive medicine is increasingly focused on characterizing oocytes and ovarian tissue composition, as well as understanding the molecular mechanisms that guide ovarian function throughout its lifecycle. High-throughput omics technologies have enabled the characterization of different molecular layers, leading to substantial advances in our understanding of their complex dynamics. However, not all molecular aspects are studied equally, and studies examining the same modalities often show inconsistencies, underscoring the need for data standardization and highlighting the potential for using transformative artificial intelligence and machine-learning (AI/ML) methods for ovary studies. OBJECTIVE AND RATIONALE: This study aims to evaluate how multi-omic studies have advanced our understanding of the ovarian lifecycle from fetal development to postmenopause. We systematically reviewed published studies that have investigated molecular/omic layers, including the genome, methylome, transcriptome, and proteome throughout ovarian development and aging. Our analysis identified key molecular and cellular patterns, highlighted inconsistencies across studies and addressed gaps in data analysis, interpretation, and reproducibility to guide future research. SEARCH METHODS: We conducted a systematic literature search of Medline (PubMed), Embase (Ovid), and Web of Science Core Collection (Clarivate) using a combination of controlled and free text terms for human ovary, oogenesis, folliculogenesis, ovary development and (epi)genome, transcriptome, proteome, and multi-omic mechanisms to find relevant articles published before August 2025. To focus the scope of the current review, studies of domesticated and farm animals, rodents and other model organisms, non-human primates, as well as those examining various human ovarian pathologies were excluded. OUTCOMES: The search identified 23 546 studies for screening, of which 637 full-text studies were assessed for eligibility. Subsequently, we extracted data from 121 studies. Most studies analyzed the transcriptome of oocytes, granulosa cells, and ovarian tissue from reproductive-age individuals (n = 91), with fewer studies examining samples from individuals of advanced reproductive age (n = 45) and fetal (n = 16) samples. Transcriptome analyses were most common (n = 103, 85%), followed by proteome (n = 19, 16%) and epigenome (n = 14, 12%) studies. We found substantial variation in how studies defined and reported participants' groups as well as in their sequencing technologies and data analysis methods, with a lack of standardized reporting of background clinical information, data analysis methods, and pipeline details. The key findings underscore the prevailing consensus on genes defining major ovarian cell types and their roles throughout the ovarian lifespan, from prenatal development to postmenopausal transformation. This review highlighted the underrepresentation of certain patient groups, particularly prepubertal and peri-/postmenopausal individuals, among researched populations, due to obvious clinical and ethical reasons. WIDER IMPLICATIONS: This scoping review offers a comprehensive overview and benchmark of the current state of high-throughput omics-based research on ovarian cellular composition and molecular dynamics. To address these shortcomings, we propose general recommendations for multi-omics ovary studies and emphasize the necessity for more thorough multi-omic data integration by effectively applying novel AI/ML approaches. They can potentially improve the quality of multi-omics analyses at both single-cell and tissue levels despite limited sample sizes and enable integration of molecular profiling data with clinical and radiology datasets, enabling a more comprehensive understanding of ovarian biology. Such advancements can enhance reproducibility of research findings and guide future research to deepen our understanding of ovarian biology and ultimately support the development of medical technologies for better preserving fertility and alleviating infertility. REGISTRATION NUMBER: A protocol was published a priori on the Open Science Framework (https://osf.io/z38gb/).

Female↗
PubMed1986

Computer-automated prediction of the mutagenicity of benzidine, 4,4"-diaminoterphenyl, 4-dimethylaminoazobenzene and 4-cyanodimethylaniline: comparison with the results of the Second UKEMS Collaborative Study.

There was agreement between the experimental results, obtained in the course of the Second UKEMS Collaborative Study, for the mutagenicity in Salmonella typhimurium of benzidine, 4,4"-diaminoterphenyl, 4-dimethylaminoazobenzene and 4-cyanodimethylaniline and the mutagenicity predicted by CASE (Computer Automated Structure Evaluation), a recently developed artificial intelligence system.

Aniline Compounds↗
PubMed1998

Application of Kohonen neural networks for the non-morphological distinction between glomerular and tubular renal disease.

BACKGROUND: A Kohonen topological map is an artificial intelligence system of the connectionist school (neural networks). The map learns the typical features of the subclasses in the learning set by means of a shortest Euclidean distance algorithm, after which self-adaptation of the neurons occurs. By its ability of self-organization and generalization, a Kohonen map is useful for pattern recognition, and its application in the medical field as an aid for decision making seems promising. This study describes the use of a Kohonen topological mapping system in the classification of renal diseases as being glomerular or tubular on basis of clinical characteristics and laboratory results. METHODS: Forty-one parameters from 75 patients were retrospectively retrieved and used to train four different Kohonen maps of 10 x 10 neurons. For reference diagnostic classification, we referred to the results of the light-microscopic examination. The classification of the patients by the four different Kohonen networks was compared to the classification by a rule-based system and by three nephrologists. We also developed a 'hybrid' decision system that makes a classification on basis of the opinion of the four networks and that of the rule-based system. RESULTS: The results show that a Kohonen map is capable of classifying the patients as having glomerular or tubular disease with a higher sensitivity and predictive value than the nephrologists and the rule-based system, and that the best classification was performed by the hybrid system: sensitivity and predictive value for the diagnosis 'glomerular' respectively 100 and 88% for the network with the most adequate results, 90 and 83% for the nephrologists, 90 and 95% for the rule-based system, and 95 and 96% for the hybrid system; sensitivity and predictive value for the diagnosis 'tubular' respectively 50 and 100% for the neural networks, 31 and 45% for the nephrologists, 81 and 68% for the rule-based system, and 87 and 82% for the hybrid system). CONCLUSION: We conclude that a Kohonen map is capable of classifying the patients as having glomerular or tubular disease with a high sensitivity and predictive value. The rule-based system performs worse than the neural networks. The most adequate results were obtained with the hybrid system.

Adult↗
PubMed1998

Use of neural networks to model complex immunogenetic associations of disease: human leukocyte antigen impact on the progression of human immunodeficiency virus infection.

Complex immunogenetic associations of disease involving a large number of gene products are difficult to evaluate with traditional statistical methods and may require complex modeling. The authors evaluated the performance of feed-forward backpropagation neural networks in predicting rapid progression to acquired immunodeficiency syndrome (AIDS) for patients with human immunodeficiency virus (HIV) infection on the basis of major histocompatibility complex variables. Networks were trained on data from patients from the Multicenter AIDS Cohort Study (n = 139) and then validated on patients from the DC Gay cohort (n = 102). The outcome of interest was rapid disease progression, defined as progression to AIDS in <6 years from seroconversion. Human leukocyte antigen (HLA) variables were selected as network inputs with multivariate regression and a previously described algorithm selecting markers with extreme point estimates for progression risk. Network performance was compared with that of logistic regression. Networks with 15 HLA inputs and a single hidden layer of five nodes achieved a sensitivity of 87.5% and specificity of 95.6% in the training set, vs. 77.0% and 76.9%, respectively, achieved by logistic regression. When validated on the DC Gay cohort, networks averaged a sensitivity of 59.1% and specificity of 74.3%, vs. 53.1% and 61.4%, respectively, for logistic regression. Neural networks offer further support to the notion that HIV disease progression may be dependent on complex interactions between different class I and class II alleles and transporters associated with antigen processing variants. The effect in the current models is of moderate magnitude, and more data as well as other host and pathogen variables may need to be considered to improve the performance of the models. Artificial intelligence methods may complement linear statistical methods for evaluating immunogenetic associations of disease.

Acquired Immunodeficiency Syndrome↗
PubMed1989

Prediction of beta-turns in proteins using neural networks.

The use of neural networks to improve empirical secondary structure prediction is explored with regard to the identification of the position and conformational class of beta-turns, a four-residue chain reversal. Recently an algorithm was developed for beta-turn predictions based on the empirical approach of Chou and Fasman using different parameters for three classes (I, II and non-specific) of beta-turns. In this paper, using the same data, an alternative approach to derive an empirical prediction method is used based on neural networks which is a general learning algorithm extensively used in artificial intelligence. Thus the results of the two approaches can be compared. The most severe test of prediction accuracy is the percentage of turn predictions that are correct and the neural network gives an overall improvement from 20.6% to 26.0%. The proportion of correctly predicted residues is 71%, compared to a chance level of about 58%. Thus neural networks provide a method of obtaining more accurate predictions from empirical data than a simpler method of deriving propensities.

Algorithms↗