SEARCH · PubMed Health

Results for “Workflow”

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 847 records · Page 47Linked to original sources

PubMed2006

A qualitative study of the implementation of a bioinformatics tool in a biological research laboratory.

OBJECTIVE: To explore how the implementation of a comprehensive new bioinformatics analysis system would affect workflow, collaboration and information management in a small genetic research lab. DESIGN: This was a longitudinal qualitative study of seven individuals involved in genomic and proteomic research. The study data were gathered using the illuminative/responsive approach of immersion in the environment. Additional qualitative data were gathered using informal semi-structured interviews, participant observation in lab meetings, and direct observation of lab researchers engaged in specific tasks. MEASUREMENTS: Interview, observation and field note data were coded and analyzed based on three analysis perspectives. A subset of the data was independently evaluated by an external researcher to enhance the trustworthiness of results. RESULTS: Three reoccurring themes were observed in the study. (1) Satisfaction and acceptance of software tools tended to be role and goal specific. (2) The system was seen primarily as a measurement system rather than a "total laboratory analysis system". (3) Lab meetings deemphasized the system, preferring more traditional data analysis techniques. These themes support the observations that the system was not used to its full potential in the lab. CONCLUSION: Themes identified in this study suggest that sophisticated genetic researchers face similar problems of technology implementation as do professionals in other fields. We recommend that leadership support and on-going training and evolution of academic curricula can improve chances of bioinformatics analysis systems becoming used more effectively.

Computational Biology↗

PubMed2007

The role of perceptions of clinicians in their adoption of a web-based antibiotic approval system: do perceptions translate into actions?

PURPOSE: Computerized Decision Support Systems have been shown to improve clinicians' performance. Clinicians' adoption of these systems is crucial for their success. Studying clinicians' perceptions can provide an insight into the determinants of clinicians' adoption of such systems. The aim of this study was to measure clinicians' perceptions of ease of use and usefulness of a web-based antibiotic approval system, and to investigate the relationship between the reported perceptions and use of the system. METHODS: Potentially identifiable coded surveys were sent to a total of 70 senior and 150 junior medical staff, and 30 pharmacists all working at a tertiary care referral teaching hospital of Melbourne, Australia. Clinicians' perceptions of ease of use and usefulness of the antibiotic approval system; clinicians' general computer use; and clinicians' usage of the antibiotic approval system were measured. RESULTS: The overall response rate from the clinicians was 53.4%. The majority of the participants (70% and above) found it easy to obtain antibiotic approval using the system. More than 80% of the participants believed that the system will decrease the inappropriate use of antibiotics at the hospital. Clinicians who were more likely to use the system also found it easy to learn (Rho=0.392, p=0.001), easy to show others how to use the system (Rho=0.298, p=0.014), easy to find additional information (Rho=0.317, p=0.009), and easy to use it within their daily workflow (Rho=0.268, p=0.028). In addition, the clinicians were also more likely to use the system if they believed that it will improve their adherence to evidence-based practice (Rho=0.352, p=0.003). CONCLUSION: The majority of clinicians in an independent investigation of the antibiotic approval system found the system easy to use and useful to them. A number of clinicians' perceptions about the system were found to be correlated with the actual usage of the system by the clinicians.

Adult↗

PubMed2004

The MEPUC concept adapts the C-arm fluoroscope to image-guided surgery.

Image-guided surgery requires surgeons to be able to manipulate the imaging modality themselves and without delay. Intraoperative fluoroscopic imaging does not meet this requirement as the C-arm fluoroscope cannot be operated or positioned by the surgeons themselves. The Motorized Exact Positioning Unit for C-arm (MEPUC) concept aims to optimize the workflow of positioning the C-arm fluoroscope. The hardware component of the MEPUC equips the fluoroscope with electric stepping motors. The software component allows the surgeon to control the fluoroscope's movements. The study presented here showed that translational movements within the x-y plane are most frequently performed when positioning the C-arm fluoroscope. Furthermore, reproducing a former projection was found to be a frequent task during image-guided procedures. In our opinion, the MEPUC concept adapts the fluoroscope to image-guided surgery. The most important improvement being definition of a bidirectional data exchange between the surgeon and the C-arm fluoroscope: positioning data from the surgeon to the C-arm fluoroscope and-subsequently-image information from C-arm fluoroscope to the surgeon.

Femoral Fractures↗

PubMed2026

A streamlined protocol for small-scale protoplast generation and CRISPR/Cpf1-mediated genome editing in Fusarium oxysporum.

Fusarium oxysporum is a significant threat to agriculture and One Health, requiring advanced molecular tools for functional genomic analyses and biological control agent development. Existing gene-editing methods are hampered by costly protoplast preparation protocols and by CRISPR-Cas9 limitations, such as restricted protospacer adjacent motif (PAM) sequences and complex guide RNA requirements. We engineered an efficient CRISPR/Cpf1 system that overcomes these issues through three main innovations: small-scale protoplast generation using filter column-based methods that greatly reduce enzyme consumption while simplifying workflows, a CRISPR/Cpf1 system with shorter guide RNA design and staggered DNA cleavage to promote homologous recombination, and minimal homology arm strategies that significantly decrease cloning complexity. Extensive validation confirms successful gene targeting with molecular verification and functional analysis via standardized pathogenicity assays. This integrated platform offers affordable, accessible tools for systematic F. oxysporum research, enhancing fundamental understanding of plant-pathogen interactions and supporting high-throughput screening vital for agricultural biotechnology and biological agent development.

CRISPR/Cpf1↗

PubMed2006

Application of a linear ion trap/orbitrap mass spectrometer in metabolite characterization studies: examination of the human liver microsomal metabolism of the non-tricyclic anti-depressant nefazodone using data-dependent accurate mass measurements.

We report herein, facile metabolite identification workflow on the anti-depressant nefazodone, which is derived from accurate mass measurements based on a single run/experimental analysis. A hybrid LTQ/orbitrap mass spectrometer was used to obtain accurate mass full scan MS and MS/MS in a data-dependent fashion to eliminate the reliance on a parent mass list. Initial screening utilized a high mass tolerance ( approximately 10 ppm) to filter the full scan MS data for previously reported nefazodone metabolites. The tight mass tolerance reduces or eliminates background chemical noise, dramatically increasing sensitivity for confirming or eliminating the presence of metabolites as well as isobaric forms. The full scan accurate mass analysis of suspected metabolites can be confirmed or refuted using three primary tools: (1) predictive chemical formula and corresponding mass error analysis, (2) rings-plus-double bonds, and (3) accurate mass product ion spectra of parent and suspected metabolites. Accurate mass characterization of the parent ion structure provided the basis for assessing structural assignment for metabolites. Metabolites were also characterized using parent product ion m/z values to filter all tandem mass spectra for identification of precursor ions yielding similar product ions. Identified metabolite parent masses were subjected to chemical formula calculator based on accurate mass as well as bond saturation. Further analysis of potential nefazodone metabolites was executed using accurate mass product ion spectra. Reported mass measurement errors for all full scan MS and MS/MS spectra was <3 ppm, regardless of relative ion abundance, which enabled the use of predictive software in determining product ion structure. The ability to conduct biotransformation profiling via tandem mass spectrometry coupled with accurate mass measurements, all in a single experimental run, is clearly one of the most attractive features of this methodology.

Antidepressive Agents↗

PubMed2004

Applications and performance of a MALDI-ToF mass spectrometer with quadratic field reflectron technology.

A new matrix-assisted laser desorption/ionization time of flight mass spectrometer (MALDI-ToF MS), developed specifically for the identification and characterization of proteins and peptides in proteomic investigations, is described. The mass spectrometer which can be integrated with the 2-D gel electrophoresis workflow is a bench-top instrument, enabling rapid, reliable and unattended protein identification in low-, as well as high-throughput proteomics applications. To obtain precise information on peptide sequences, the instrument utilizes a timed ion gate and a unique quadratic field reflectron (Z2 technology), allowing single-run, post-source decay (PSD) of selected peptides. In this study, the performance of the instrument in reflectron, PSD and linear mode, respectively, was investigated. The results showed that the limit of detection for a single peptide in reflectron mode was 125 amol with a signal to noise ratio exceeding 20. Average mass resolution for peptides larger than 2000 u was around 13,000 full width, half maximum (FWHM). The limit for protein identification during peptide mass fingerprinting (PMF) was 500 amol with a sequence coverage of 18%. Mass error during PMF analysis was less than 15 ppm for 17 out of 25 (68%) identified peptides. In PSD mode, a complete series of y-ions of a CAF-derivatized peptide could be obtained from 3.75 fmol of material. The average mass error of PSD-generated fragments was less than 0.14 u. Finally, in linear mode, intact proteins with molecular masses greater than 300,000 u were detected with mass errors below 0.2%.

Electrophoresis, Gel, Two-Dimensional↗

PubMed2026

Functional screening of ZIP8 naturally occurring variants identifies pathogenic mutations and trafficking defects.

The rapid expansion of human genomic data has revealed a large number of naturally occurring variants, creating a major challenge for functional annotation. The human metal transporter SLC39A8 (ZIP8) is a clinically important divalent metal transporter, yet most of its documented variants remain uncharacterized. Here, we developed a workflow to functionally evaluate ZIP8 variants by integrating laser ablation inductively coupled plasma time-of-flight mass spectrometry (LA-ICP-TOF-MS) with scaled-up cell-based transport assays. Using this method, we systematically analyzed 33 naturally occurring missense variants located in the extracellular domain (ECD) of ZIP8. The assay enables direct quantification of intracellular metal accumulation with substantially improved throughput (∼150 samples per hour). Functional screening identified 14 potential pathogenic variants with significantly reduced transport activity. Comparison with computational predictions revealed a moderate correlation between activity and AlphaMissense pathogenicity scores (R2 = 0.423), while an error rate of ∼20% for AlphaMissense underscores the need for experimental validation. Flow cytometry analysis showed that most loss-of-function variants exhibit impaired trafficking of the protein to the cell surface possibly due to mutation-caused protein misfolding or instability. Structural mapping of activity-compromised variants, together with functional assessment of the ZIP8-ECD, highlights the importance of this domain in ZIP8 expression and intracellular protein trafficking. Together, this work establishes a scalable approach for functional screening of metal transporter variants and provides new insights into the structure-function relationships of ZIP8.

Journal Article↗

PubMed2004

Incorporating ideas from computer-supported cooperative work.

Many information systems have failed when deployed into complex health-care settings. We believe that one cause of these failures is the difficulty in systematically accounting for the collaborative and exception-filled nature of medical work. In this methodological review paper, we highlight research from the field of computer-supported cooperative work (CSCW) that could help biomedical informaticists recognize and design around the kinds of challenges that lead to unanticipated breakdowns and eventual abandonment of their systems. The field of CSCW studies how people collaborate with each other and the role that technology plays in this collaboration for a wide variety of organizational settings. Thus, biomedical informaticists could benefit from the lessons learned by CSCW researchers. In this paper, we provide a focused review of CSCW methods and ideas-we review aspects of the field that could be applied to improve the design and deployment of medical information systems. To make our discussion concrete, we use electronic medical record systems as an example medical information system, and present three specific principles from CSCW: accounting for incentive structures, understanding workflow, and incorporating awareness.

Artificial Intelligence↗

PubMed2005

Using qualitative studies to improve the usability of an EMR.

The adoption of electronic medical records (EMRs) and user satisfaction are closely associated with the system's usability. To improve the usability of a results management module of a widely deployed web-based EMR, we conducted two qualitative studies that included multiple focus group and field study sessions. Qualitative research can help focus attention on user tasks and goals and identify patterns of care that can be visualized through task modeling exercises. Findings from both studies raised issues with the amount and organization of information in the display, interference with workflow patterns of primary care physicians, and the availability of visual cues and feedback. We used the findings of these studies to recommend design changes to the user interface of the results management module.

Attitude to Computers↗

PubMed2005

Human-centered design of a distributed knowledge management system.

Many healthcare technology projects fail due to the lack of consideration of human issues, such as workflow, organizational change, and usability, during the design and implementation stages of a project's development process. Even when human issues are considered, the consideration is typically on designing better user interfaces. We argue that human-centered computing goes beyond a better user interface: it should include considerations of users, functions and tasks that are fundamental to human-centered computing. From this perspective, we integrated a previously developed human-centered methodology with a Project Design Lifecycle, and we applied this integration in the design of a complex distributed knowledge management system for the Biomedical Engineer (BME) domain in the Mission Control Center at NASA Johnson Space Center. We analyzed this complex system, identified its problems, generated systems requirements, and provided specifications of a replacement prototype for effective organizational memory and knowledge management. We demonstrated the value provided by our human-centered approach and described the unique properties, structures, and processes discovered using this methodology and how they contributed in the design of the prototype.

Artificial Intelligence↗

PubMed2003

Qualitative and quantitative proteomics by two-dimensional gel electrophoresis, peptide mass fingerprint and a chemically-coded affinity tag (CCAT).

The chemically-coded affinity tag (CCAT) method combines standard electrophoresis protocols with MALDI-TOF-MS analysis to identify and quantify protein abundances in complex samples in one step. This method is designed to fit into the workflow of SDS-PAGE or two-dimensional electrophoresis (2-DE) only requiring basic proteome laboratory equipment. Prior to electrophoresis two protein samples are separately labelled with a heavy or a light version of the CCAT reagent via reduced cysteines in the proteins. Equal amounts are then combined and electrophoretically separated. Proteins can then be excised from the gel to obtain their peptide mass fingerprint by mass spectrometry. This fingerprint enabled not only identification, but also quantification by comparing relative peak intensities of CCAT-labelled peptides. In this article, we display how the CCAT method can be used to analyse two protein samples in one gel and that the peak intensities of labelled peptides reflect the abundance of a protein in it.

Affinity Labels↗

PubMed2026

A genome-wide coverage-based pipeline for the identification of host-derived candidate DNA biomarkers from cell-free blood.

We have created a new data-analysis pipeline for the discovery of host-specific candidate DNA biomarkers derived from sequencing data of cell-free blood. Unlike approaches that rely on specific molecular or genetic signatures, our method leverages the coverage distribution of cell-free DNA sequences mapped to a reference genome, applying statistical analyses to identify informative short genomic regions for biomarker discovery. The pipeline is applicable to diverse diseases and can be used to analyze cell-free DNA sequences from plasma or serum to identify candidate biomarkers that are characteristic of disease states in mammals. Core functionalities were developed in Java and integrated with open-source software tools for the preprocessing of raw sequencing data, complemented by Python scripts for the machine-learning analysis and statistical validation. The pipeline is designed for HPC use and users can access the pipeline through a Galaxy workflow, which offers a user-friendly web interface for input selection prior to execution and analysis progress monitoring. Performance tests, carried out using duplicate sets of COVID-19 samples and controls, showed linear scalability of execution time with an increasing dataset size, as well as a substantial reduction in execution time through parallelized computation, whereby each HPC node is used to process the data of one chromosome. Further statistical tests confirmed the quality of the pipeline's results by showing that the set of identified candidate biomarkers remained stable across varying dataset sizes.

Biomarkers↗

PubMed2025

Foundations of Artificial Intelligence in Hepatology: What a Clinician Needs to Know.

This review focuses on foundational knowledge about artificial intelligence (AI) in hepatology, exploring how AI, including machine learning and deep learning, leverages large-scale clinical data to transform the diagnosis, risk assessment, prognostication, and management of liver diseases. Online resources are described to offer fundamental AI knowledge and essential technical skills and to facilitate clinician participation across the entire AI lifecycle, ensuring they contribute not only as end users but also in development and deployment. Unlike traditional statistical approaches that prioritize interpretable parameters and clinical insight, AI focuses on maximizing predictive accuracy by identifying complex, often non-linear patterns using high-dimensional data, albeit often at the cost of model interpretability. AI is demonstrating clinical utility in liver histopathology and radiological imaging, significantly improving detection accuracy for cirrhosis, clinically significant portal hypertension, and hepatocellular carcinoma. Beyond diagnostics, AI-driven prediction models are emerging to provide personalized risk stratification for the development of liver-related complications and treatment guidance, based on complex data including longitudinal laboratory results, comorbidities, and co-medication use to monitor disease progression and therapy response. The field is rapidly expanding into novel areas such as analyzing patient-reported outcomes, genomic data, and real-time liver function monitoring, offering deeper mechanistic insights alongside clinical tools. Despite the potential to revolutionize hepatology practice and research, successful integration into routine care faces challenges. These include seamless workflow integration with existing electronic health records, establishing clear liability frameworks, and guaranteeing protection of patient privacy. Addressing these hurdles requires collaborative efforts from clinicians, researchers, and regulators to develop best practices and governance. Understanding the transformative capabilities, current applications, emerging frontiers, and essential implementation considerations is crucial for clinicians navigating the evolving AI landscape and responsibly utilizing its power for improved patient outcomes.

PROBAST+AI↗

PubMed2026

Interactive gaming during inhalational induction of anesthesia reduces pediatric patient anxiety and improves induction compliance: A randomized controlled trial.

BACKGROUND: Preoperative anxiety affects up to 75% of pediatric surgical patients and is associated with adverse postoperative outcomes. Traditional anxiolytic strategies with premedication carry drawbacks including delayed recovery and paradoxical reactions, driving interest in non-pharmacologic alternatives. Audiovisual distraction represents one approach, encompassing passive methods (e.g. watching a video) and active modalities (e.g. interactive gaming). The Bedside Entertainment and Relaxation Theater (BERT) is a projection-based environment that enables audiovisual distraction during induction. Whether BERT-based interactive gaming reduces anxiety and improves induction compliance compared to standard perioperative care remains unknown. METHODS: This single-center RCT enrolled 74 pediatric patients aged 4 to 14 undergoing inhalational induction, randomized to standard care (SOC) or interactive gaming via BERT added to SOC during induction (BERT). The primary outcome was change in patient anxiety from baseline to induction, measured using the Modified Yale Preoperative Anxiety Scale (mYPAS). Secondary outcomes included caregiver anxiety, induction compliance, OR efficiency, opioid administration, and OR staff perceptions. RESULTS: Patients in the BERT group experienced significantly smaller increases in anxiety from baseline to induction than SOC (median mYPAS increase [IQR]: 0 [0 to 0] vs 10 [0 to 38], p < 0.001). Induction compliance improved, with lower Induction Compliance Checklist (ICC) scores indicating fewer induction-related disruptive behaviors than SOC (median: 0 vs 1, shift -1 [95% CI: -2 to 0]; p = 0.004). Caregiver anxiety increased less in the BERT group than SOC (mean STAI increase: 0.18 vs 1.89, difference - 1.7, [95% CI: -3.2 to -0.26]; p = 0.022). OR staff reported high acceptability, with 97% supporting continued use. No differences were observed in OR efficiency or opioid administration between groups. CONCLUSIONS: Interactive gaming via BERT attenuated increases in patient anxiety, improved induction compliance, and reduced the rise in caregiver anxiety without prolonging OR time. These findings support BERT as an effective, workflow-compatible anxiolytic strategy for pediatric inhalational induction.

Humans↗

PubMed2005

Comparative evaluation of the Cobas Amplicor HIV-1 Monitor Ultrasensitive Test, the new Cobas AmpliPrep/Cobas Amplicor HIV-1 Monitor Ultrasensitive Test and the Versant HIV RNA 3.0 assays for quantitation of HIV-1 RNA in plasma samples.

BACKGROUND: There are several commercially available assays for the quantitation of HIV RNA. A new automated specimen preparation system, the Cobas AmpliPrep, was developed to automate this last part of the PCR. OBJECTIVES AND STUDY DESIGN: We compared the results obtained by the Roche Cobas Amplicor HIV-1 Monitor Ultrasensitive Test (MCA, manual sample preparation) with those by the Versant HIV-1 RNA 3.0 assay (bDNA). Secondly we compared the MCA with the new Cobas AmpliPrep/Cobas Amplicor HIV Monitor Ultrasensitive Test (CAP/CA, automated specimen preparation) by investigating clinical patient samples and a panel of HIV-1 non-B subtypes. Furthermore, we assessed the assay throughput and workflow (especially hands-on time) for all three assays. RESULTS: Seventy-two percent of the 140 investigated patient samples gave concordant results in the bDNA and MCA assays. The MCA values were regularly higher than the bDNA values. One sample was detected only by the MCA within the linear range of quantification. In contrast, 38 samples with results <50 copies/ml in the MCA showed in the bDNA results between 51 and 1644 copies/ml (mean value 74 copies/ml); 21 of these specimens were shown to have detectable HIV RNA < 50 copies/ml in the MCA assay. The overall agreement between the MCA and the CAP/CA was 94.3% (551/584). The quantification results showed significant correlation, although the CAP/CA generated values slightly lower than those generated by the manual procedure. We found that the CAP/CA produced comparable results with the MCA test in a panel of HIV-1 non-B subtypes. CONCLUSIONS: All three assays showed comparable results. The bDNA provides a high sample throughput without the need of full automation. The new CAP/CA provides reliable test results with no HIV-subtype specific influence and releases time for other works in the laboratory; thus it is suitable for routine diagnostic PCR.

Automation↗

PubMed2005

Database management systems for process safety.

Several elements of the process safety management regulation (PSM) require tracking and documentation of actions; process hazard analyses, management of change, process safety information, operating procedures, training, contractor safety programs, pre-startup safety reviews, incident investigations, emergency planning, and compliance audits. These elements can result in hundreds of actions annually that require actions. This tracking and documentation commonly is a failing identified in compliance audits, and is difficult to manage through action lists, spreadsheets, or other tools that are comfortably manipulated by plant personnel. This paper discusses the recent implementation of a database management system at a chemical plant and chronicles the improvements accomplished through the introduction of a customized system. The system as implemented modeled the normal plant workflows, and provided simple, recognizable user interfaces for ease of use.

Database Management Systems↗

PubMed2026

In silico identification of DNMT1 inhibitors from the PlantCyc database through computational approach to assess the anti-cancer potential of nutraceutical compounds in breast cancer.

Breast cancer accounts for a disproportionate share of global cancer-related deaths, with 670,000 fatalities and 2.3 million new diagnoses recorded in women during 2022 alone. Existing treatment modalities carry considerable toxicity burdens, and resistance to available agents remains an unresolved clinical problem. DNA methyltransferase 1 (DNMT1), the enzyme chiefly responsible for maintaining genome-wide methylation patterns during DNA replication, has been mapped out as a high-value target in breast cancer because its dysregulation silences tumour suppressor genes through promoter hypermethylation. The present work involves hierarchical in silico workflow to screen 4549 plant-derived compounds from the PlantCyc database (v16.0.3) against the human DNMT1 catalytic domain (PDB ID: 4WXX). Ten top-scoring compounds were taken forward for molecular docking via AutoDock Vina; Quercetin and Kaempferol both recorded the highest binding affinities at -9.5 kcal/mol, Wogonin (-9.3 kcal/mol) and Xanthohumol (-8.1 kcal/mol) also emerged as strong binders. Pharmacokinetic evaluation using ADMET-AI confirmed that all 10 compounds met Lipinski's rule of five, with human intestinal absorption values at or above 0.98. Wogonin and Xanthohumol were selected for a 100 ns all-atom molecular dynamics (MD) simulation in GROMACS due to their well-rounded ADMET profiles and limited existing data on their specific interactions with DNMT1 in breast cancer. Across all measured trajectory metrics, backbone RMSD, residue fluctuation, radius of gyration, solvent-accessible surface area, and intermolecular hydrogen bond count, Wogonin formed a more stable, compact complex. These findings suggest that Wogonin and Xanthohumol are non-toxic nutraceutical candidates suitable for DNMT1 targeted epigenetic therapy, with computational foundation strong enough to facilitate future in vitro and in vivo validation work.

Humans↗

PubMed2026

Artificial intelligence (AI) uses in stereotactic radiosurgery (SRS): diagnosis with brain metastasis (BM) - A systematic review.

BACKGROUND: Brain metastases (BM) are the most common intracranial tumors in adults, and stereotactic radiosurgery (SRS) has become a mainstay of management. However, several diagnostic challenges persist in the SRS pathway, particularly the differentiation of radiation necrosis (RN) from true tumor progression, which conventional MRI and even advanced imaging techniques often cannot reliably resolve. Recent advances in artificial intelligence (AI) offer the potential to address these diagnostic limitations. This systematic review synthesizes current literature on AI applications for MRI-based diagnostic decision support in BM patients undergoing SRS, with a focus on radiomics and deep learning tools for distinguishing RN from progression, classifying molecular and histologic subtypes, and predicting treatment response. METHODS: A systematic review was performed in accordance with PRISMA guidelines. PubMed, Web of Science, and Scopus were searched using a targeted query combining terms related to AI, brain metastasis, diagnosis or imaging, and SRS. After screening 483 records and applying strict inclusion and exclusion criteria, 18 studies published between 2015 and 2025 were included. Data were extracted on study design, cohort characteristics, imaging modality, AI methodology, validation strategy, and reported diagnostic performance. RESULTS: Among the 18 included studies, AI models demonstrated strong performance across diagnostic tasks in the BM-SRS pathway. The differentiation of RN from true tumor progression was the most extensively studied application, addressed by 14 of 18 studies, with reported AUCs ranging from 0.71 to 0.94. Support vector machines, random-forest ensembles, convolutional neural networks, and transformer-based multimodal architectures were widely used. The literature evolved from single-sequence radiomic classifiers in 2018 to multimodal deep learning frameworks fusing imaging with clinical and genomic data in 2025. Contrast-enhanced T1-weighted MRI was the dominant imaging input, and texture-based radiomic features (GLCM, GLSZM, GLDM, and wavelet-derived features) were the most consistently predictive. The highest-performing models reached AUCs of 0.85-0.91 through multimodal integration of imaging with clinical and genomic features, and consistently outperformed expert neuroradiologist read on matched cases. Remaining studies addressed longitudinal segmentation-based detection of local failure and adverse radiation effects, BRAF mutation status in melanoma BM, early Gamma Knife treatment response, and primary tumor histology classification, with more variable performance. CONCLUSION: AI models, particularly those integrating MRI-derived radiomic features with clinical and genomic data, show high accuracy in supporting diagnostic decisions for BM patients treated with SRS. The post-SRS differentiation of radiation necrosis from true tumor progression has reached the greatest level of maturity and is closest to clinical translation, with potential to reduce unnecessary biopsies, personalize surveillance intervals, and rationalize treatment-pathway decisions. Other diagnostic applications, including molecular subtyping and primary tumor histology classification, remain exploratory and require further multicenter validation. Integration of AI tools into multidisciplinary tumor-board workflows, combined with prospective validation and standardized reporting, will be essential to realize the full clinical benefits of AI in SRS for brain metastases.

Humans↗