SEARCH · Search PubMed

Results for “Clinical decision rules”

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.

SourceResearch connectionFrom yearThrough year

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

1,385 records · Page 4Linked to original sources

PubMed2026

How AI-supported intelligent systems support infection prevention and control training in healthcare: A systematic review of educational functions and outcomes.

AIMS: Artificial intelligence (AI)-supported intelligent systems have been increasingly incorporated into infection prevention and control (IPC) education and training, primarily to support the monitoring of observable behaviors and the provision of feedback. However, existing evidence has focused largely on short-term compliance outcomes, with limited synthesis of the educational role of AI-supported intelligent systems in supporting sustained IPC competence. This systematic review examined how AI-supported intelligent systems have been designed and used to support IPC education and training, with a focus on system characteristics, educational functions, and reported outcomes. DESIGN: A systematic literature search was conducted across the PubMed/MEDLINE, Embase, Cochrane, and CINAHL databases. DATA SOURCES: A total of 18 studies met the inclusion criteria. Findings were qualitatively synthesized according to system design characteristics, educational functions, and outcome domains. REVIEW METHODS: Methodological quality was appraised using the Mixed Methods Appraisal Tool. RESULTS: Most AI-supported intelligent systems focused on hand hygiene and relied on fully automated monitoring systems to capture behaviors and provide performance feedback. Educational functions were predominantly limited to performance assessment, automated feedback, and reminders. Outcomes were mainly measured using compliance or performance metrics, whereas sustained behavioral change and decision quality were rarely assessed. CONCLUSIONS: AI-supported intelligent systems have been used primarily to reinforce short-term IPC performance and compliance. However, their current applications for supporting sustained competence over time remain limited. The findings of this review suggest that AI-supported intelligent systems may serve as maintenance-oriented educational support by extending learning beyond initial instruction through repeated practice and feedback. Future research should prioritize outcome measures that capture the durability of performance and decision-making processes to better align AI-supported intelligent systems used in IPC education and training with the educational demands of clinical practice.

Humans↗

PubMed2026

Molecular Diagnostics for WHO Priority Bacterial Pathogens: A Bibliometric Mapping of Diagnostic Platforms, Resistance Markers, and Antimicrobial Resistance Research Trends.

Antimicrobial resistance (AMR) constrains effective treatment and carries implications for infection control, surveillance, and public health. The World Health Organization (WHO) priority bacterial pathogen framework has intensified the need for diagnostic innovation by redefining research priorities around organisms combining high disease burden with complex resistance profiles. Molecular diagnostics have accordingly moved beyond culture-based workflows, integrating rapid pathogen identification, resistance-marker detection, genomic surveillance, and clinical decision support. The present study conducted a bibliometric mapping of the literature on WHO priority pathogens. Rather than addressing resistance at a general level or a single pathogen or technology, it integrates priority pathogens, molecular platforms, and resistance markers within a single framework, tracing their joint thematic and temporal evolution along an explicit pathogen-platform-marker axis. Scopus-indexed articles and reviews (2000-2025) were retrieved, yielding 1746 publications after screening adapted from the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Analyses used Bibliometrix/Biblioshiny, R, and VOSviewer. The literature expanded markedly after 2018, led by China and the United States. Methicillin-resistant Staphylococcus aureus (MRSA), Mycobacterium tuberculosis, Enterococcus faecium, and the Enterobacterales-carbapenemase axis constituted the principal thematic cores, whereas conventional polymerase chain reaction (PCR)/nucleic acid amplification testing (NAAT) and whole-genome sequencing were the dominant platforms. Overall, the field has evolved from pathogen detection into an AMR-centered translational domain encompassing resistance prediction, genomic epidemiology, surveillance, and clinical decision support. Diagnostic development, stewardship, and surveillance depend on hybrid workflows coupling rapid marker-targeted assays with genome-based characterization, delivering actionable resistance within clinically meaningful timeframes, and extending coverage to underrepresented pathogens and platforms.

Humans↗

PubMed2026

Views and Experiences of People With Dementia, Informal Caregivers and Professionals on Eating and Drinking Difficulties: A Qualitative Systematic Review.

AIM: This study aims to explore the views and experiences of people with dementia, informal caregivers and professionals regarding eating and drinking difficulties. DESIGN: A qualitative systematic review was conducted. METHODS: The Preferred Reporting Items for Systematic Reviews and Meta-analysis guidelines were used to conduct this systematic review. The quality of the included studies was assessed using the Joanna Briggs Institute Critical Appraisal Checklist for Qualitative Research, and the data were thematically synthesised using Thomas and Harden's three-stage method. DATA SOURCES: Six electronic databases (PubMed, EMBASE, Cochrane Library, Web of Science, CINAHL and PsycINFO) were searched from their respective inception dates to August 2025 to identify relevant studies. RESULTS: Thematic analysis of the 16 included studies identified four key themes: (1) Physiological and psychological changes in people with dementia and caregivers; (2) factors influencing eating and drinking in people with dementia; (3) needs and recommendations for people with dementia, informal caregivers and professionals; (4) selection of eating methods for end-stage people with dementia. CONCLUSIONS: Eating and drinking difficulties affect the well-being of both patients and caregivers. A good dining environment improves mealtime pleasure but demands caregivers' time and energy. All parties emphasised the importance of effective communication. In end-stage dementia, professional assistance is crucial for enteral nutrition decisions. IMPLICATIONS FOR THE PROFESSION AND/OR PATIENT CARE: Collaboration among patients, caregivers and professionals is vital for creating tailored nutritional plans and improving mealtime environments, thereby enhancing nutritional intake. In advanced dementia, providers must provide balanced information on comfort feeding versus enteral nutrition to aid decision-making. IMPACT: What problems were addressed in this study? This study addressed the lack of a consolidated, tri-perspective understanding of eating and drinking difficulties in dementia care settings. What are the main findings? Four key themes were identified: physiological and psychological changes, influencing factors, stakeholder needs and end-of-life decision-making. Where and on whom will the research have an impact? This will impact care practices for people with dementia and inform the training and support of informal caregivers and healthcare professionals.

Humans↗

PubMed2026

Privacy, security, and reliability risks of artificial intelligence in healthcare: a systematic review of empirical evidence.

BACKGROUND: Artificial intelligence (AI) is increasingly integrated into healthcare information systems, supporting clinical decision-making, imaging analysis, and predictive modeling. While these applications offer operational and clinical benefits, they also introduce emerging risks to patient privacy, data security, and system reliability. OBJECTIVE: To systematically review empirical evidence on privacy breaches, security vulnerabilities, and misuse associated with AI applications in healthcare settings. METHODS: PubMed, Embase, Web of Science, Scopus, IEEE Xplore, and ACM Digital Library were searched for empirical studies published between January 2015 and November 2025 that evaluated AI use or misuse in clinical diagnosis, treatment, or decision-making. Two reviewers independently screened studies and extracted data using a standardized form. Findings were synthesized narratively due to heterogeneity in study designs, AI methods, and reported outcomes. RESULTS: Of 7,285 records identified through database searches and 205 through citation screening, 22 empirical studies met the inclusion criteria, spanning multiple clinical domains and data modalities, predominantly medical imaging applications. Five recurring threat categories were identified: patient re-identification, membership inference, unauthorized access and adversarial exploitation, input manipulation, and misuse or overinterpretation of AI outputs. Across studies, AI models were shown to encode latent biometric signals across diverse data types, limiting the effectiveness of traditional anonymization and synthetic data approaches. Adversarial attacks and input manipulation were also shown to compromise diagnostic performance and system integrity. CONCLUSION: This systematic review provides empirical evidence suggesting that contemporary AI systems in healthcare introduce privacy and security risks that may challenge traditional assumptions about data protection. These findings underscore the need for privacy- and security-by-design approaches and governance frameworks that address risks across the AI lifecycle.

Humans↗

PubMed2026

Toward personalized interventions for preventing depression in primary care: Qualitative and quantitative findings from the e-predictD pilot study.

BACKGROUND: The predictD intervention, delivered by family physicians (FPs), has demonstrated effectiveness and cost-efficiency in preventing depression and anxiety. The e-predictD study aims to design, develop, and evaluate a novel personalized intervention for depression prevention by integrating information and communication technologies (ICTs), risk prediction algorithms, and decision support systems (DSS) for both patients and FPs. OBJECTIVE: To evaluate the satisfaction, usability, and acceptability, of a beta version of the e-predictD intervention in primary care settings. METHODS: The e-predictD intervention follows a biopsychosocial approach, including an initial patient-FP interview, specific FP training, and an app. A β-version was tested in a pilot study without a control group over three months. The app integrates a validated depression risk prediction algorithm, decision algorithms, and a monitoring system supporting the DSS. The DSS generates a personalized prevention plan (PPP) from eight intervention modules: physical exercise, social relationships, problem-solving, communication skills, decision-making, assertiveness, sleep improvement, and cognitive restructuring. Patients and FPs discussed the PPP in a 15-minute baseline interview, selecting modules for implementation over three months. Semi-structured interviews gathered feedback. Assessments included depression (PHQ-9), anxiety (GAD-7), quality of life (SF-12), and major depression risk (predictD algorithm). RESULTS: Six FPs from six Spanish cities enrolled 56 non-depressed patients at moderate-to-high risk of depression; 47 (84%) completed follow-up. The app was used for a median of six days (interquartile range: 1-30). Both FPs and patients expressed satisfaction, leading to incorporated improvements. After three months, significant reductions in major depression risk and anxiety symptoms were observed, alongside improved mental quality of life. However, no significant changes were found in depressive symptoms or physical quality of life. CONCLUSION: This pilot study supports the feasibility and acceptability of the e-predictD β-version, despite lower-than-expected app usability. Health improvements were observed, warranting confirmation in a randomized controlled trial. TRIAL REGISTRATION: ClinicalTrials.gov NCT03990792.

Adult↗

PubMed2026

Artificial intelligence for anticancer drug discovery from natural products of macroalgae and sponges: A systematic review.

Marine natural products (MNPs) from macroalgae and marine sponges have inspired clinically important anticancer agents, including the cytarabine pharmacophore and the eribulin scaffold, while cyanobacterial dolastatin chemistry supplies the auristatin payloads of several marine-inspired antibody-drug conjugates (ADCs) such as brentuximab vedotin. Artificial intelligence (AI) methods, encompassing both classical machine learning (ML) with hand-engineered features and modern deep learning (DL) with many-layered neural networks, are increasingly supporting key decisions in natural-product anticancer drug discovery, including bioactivity prediction, target identification, absorption, distribution, metabolism, excretion and toxicity (ADMET) filtering, generative analogue design, and the selection of preclinical candidates. DL architectures relevant to this field include graph neural networks, transformer-based molecular generators, diffusion models for protein-ligand docking, and convolutional networks for mass spectrometry, while classical ML contributes interpretable fingerprint-based bioactivity models and molecular networking for dereplication. This review follows a systematic literature review methodology to organize the landscape of AI methods now applied to MNP anticancer discovery, distinguishing ML and DL approaches where relevant, situating them within the chemical context of macroalgal and sponge-derived oncology leads, and critically examining published case studies, including validation level (computational, in vitro, in vivo, clinical). The principal bottleneck for medical translation has shifted partly from algorithmic capability toward data infrastructure and experimental validation. Sparse, heterogeneous, and taxonomically biased bioactivity records limit what current models can learn and reduce the reliability of AI-prioritized candidates entering the preclinical pipeline. A roadmap is proposed that prioritizes open MNP-specific benchmarks, symbiont-aware modeling, and active learning loops with synthesizability and ADMET constraints. These AI workflows may accelerate the prioritization of marine-derived anticancer leads and support earlier, more evidence-based translational decisions in oncology drug development.

Biological Products↗

PubMed2026

Meningioma methylation profiling as a complement to WHO grading: a single-center experience.

OBJECTIVE: The methylation profile of meningiomas is a promising predictive tool that may improve risk stratification beyond WHO grading. This study aimed to evaluate the clinical relevance and real-world applicability of routine epigenetic testing in meningioma management. METHODS: The authors retrospectively analyzed patients who underwent meningioma resection between January 2021 and December 2023. Histopathological grading (WHO 2021) and methylation profiling (methylation class [MC]) with the MethylationEPIC v1.0 (850k) chip were performed by an independent neuropathologist. RESULTS: A total of 106 patients were included; 81 tumors (76%) were classified as WHO grade 1, 20 (19%) as grade 2, and 5 (5%) as grade 3. Epigenetically, 55 tumors (52%) were classified as benign, 18 (17%) as intermediate, and 2 (2%) as malignant; 31 (29%) could not be classified. Discordances between WHO grading and methylation profiling were observed in 18 of 74 cases. Tumor board decisions were made after a median of 8 days postoperatively, guided by WHO grading; however, the epigenetic report was only available after a median of 23 days. During follow-up, 20 patients experienced tumor progression. Progression was significantly associated with the MC (r = -0.4, p < 0.001) and tumor volume (r = 0.4, p = 0.0005), but not with WHO grading (r = 0.17, p = 0.084). However, the relatively high rate of unclassified tumors and delayed result availability limited the direct impact of MC profiling on immediate clinical decision-making. Interestingly, progression-free survival in MC-unclassified tumors mirrored that of the intermediate group. CONCLUSIONS: Methylation profiling demonstrates superior predictive accuracy for meningioma progression and complements WHO grading, especially in identifying malignant meningiomas. However, its current clinical utility is constrained by technical and logistical limitations. In real-world practice, epigenetic classification should therefore be considered a complementary tool rather than a replacement for established histopathological assessment.

Humans↗

PubMed2026

Blinding integrity in psychedelic research: Evidence from a comparative randomized controlled trial of psilocybin, MDMA, and methylphenidate in healthy volunteers.

Maintaining effective blinding is a major methodological challenge in psychedelic research. This study provides a comprehensive evaluation of blinding integrity in 120 healthy volunteers who received either psilocybin, MDMA, or methylphenidate (active placebo) in a double-blind, randomized controlled trial. Using a multi-level assessment incorporating forced-choice substance guesses, certainty ratings, decision factors, and subjective substance effects, the analyses characterize blinding integrity and its relation to the substance experience. Results indicate that overall blinding was insufficient, with psilocybin showing the highest rates of functional unblinding, MDMA moderate levels, and methylphenidate the lowest. As an active placebo, methylphenidate provided more effective blinding for MDMA than for psilocybin. Incorporating certainty levels of substance guesses revealed a more differentiated pattern, with lower functional unblinding rates. Decision factors and subjective substance experiences were associated with phenomenological substance effects. Prior substance experiences did not influence accuracy of forced-choice substance guesses. These findings provide empirical guidance for the design and reporting of blinding procedures in psychedelic trials and underscore the value of systematic, multi-level assessment of blinding integrity.

Humans↗

PubMed2026

Prevalence and Factors Associated with Receiving a Prescription for a Direct Oral Anticoagulant Among Patients with Atrial Fibrillation on Hospice Admission.

Atrial fibrillation (AF) is prevalent in hospice care, but anticoagulation decisions in this population are not well understood. In this cross-sectional study, we described the prevalence and characteristics associated with direct oral anticoagulant (DOAC) prescription on hospice admission. We used electronic health data from adult decedents with AF in a large, for-profit hospice chain in the United States between January 1, 2017 and December 31, 2019. We used multivariable logistic regression with results reported as adjusted odds ratios (AORs) and 95% confidence intervals (CIs). Among 13,233 decedents, mean (standard deviation [SD]) age was 84.2 (9.9) years, 53.6% were female, 65.1% were White, and 56.1% were referred to hospice from a hospital. Mean (SD) CHA2DS2-VASc score were 3.8 (1.4) for males and 4.8 (1.3) for females, and mean (SD) HAS-BLED score was 2.2 (1.0). Overall, 8% of patients received a DOAC prescription on hospice admission. Characteristics associated with receiving a DOAC prescription included PPS scores of ≥ 20% (compared to scores < 20%), and receiving hospice care at home, nursing home, assisted living facility, or residential care home (compared to inpatient hospice). Further studies about the risks and benefits of DOAC use are needed to optimize decision-making in this population.

DOAC↗

PubMed2026

Criteria for Safe Hospital Discharge in Bronchiolitis: A Systematic Review.

Bronchiolitis is the leading cause of hospital presentation and admission for infants in Australasia. We aimed to synthesise current evidence on the effect of discharge criteria for infants (aged < 12 months) who are presenting to or are admitted to hospital with bronchiolitis, to inform a binational guideline recommendation update. Systematic searches were conducted on MEDLINE, EMBASE, PubMed, Cochrane Library and CINAHL (last search 19 February 2025) for non-randomised studies evaluating hospital discharge criteria in bronchiolitis. The primary outcomes were length of stay (LOS) and readmission rates. The risk of bias (ROBINS-I) and certainty of the evidence (GRADE) were appraised, and findings were narratively synthesised. GRADE evidence-to-decision methodology, expert consensus voting and interest-holder consultation were used to finalise the recommendation update. Two retrospective observational studies were included (N = 2697) (low to very low quality), reporting on unique discharge criteria. In both studies, use of the discharge criteria was associated with a significant reduction in LOS relative to alternative protocols. There was no significant difference in readmission rates observed in either study. There was low to very low certainty evidence across outcomes due to risk of bias, indirectness and imprecision. The review findings informed a recommendation update for safe discharge criteria in the 2025 Australasian Bronchiolitis Guideline update. Updated, prescriptive discharge criteria and flow chart were developed, covering clinical stability, oxygen saturation/support, feeding difficulties, caregiver confidence and education on deterioration, social factors and follow-up. The revised criteria provide clinicians with increased certainty in decision-making in bronchiolitis, albeit with further research needed.

Humans↗

PubMed2026

Advancing nursing education through social and emotional learning: A systematic review guided by the Collaborative for Academic, Social, and Emotional Learning framework.

BACKGROUND: With Generation Z entering the nursing workforce in growing numbers, strengthening social and emotional learning is critical for academic success, professional adaptation, and safe practice. However, the existing evidence remains fragmented because of varied interventions and inconsistent approaches. OBJECTIVES: This systematic review examined (1) the social and emotional learning essential for nursing students and nurses within the Collaborative for Academic, Social, and Emotional Learning framework, (2) their impact on educational and clinical outcomes, and (3) implications for advancing nursing education and practice. METHODS: Following Joanna Briggs Institute methodology and Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, five international (PubMed, EMBASE, CINAHL, PsycINFO, Cochrane) and three Korean (RISS, KoreaMed, KMBASE) databases were searched up to June 2025. Eighteen studies involving 2,952 participants met the inclusion criteria, including quasi-experimental quantitative studies, descriptive quantitative studies, qualitative studies, and mixed-methods studies. The methodological quality of the included studies was appraised using the Mixed Methods Appraisal Tool. RESULTS: Within the Collaborative for Academic, Social, and Emotional Learning framework, relationship skills and self-management were the most frequently studied competencies, emphasizing teamwork, communication, and stress regulation. Self-awareness and social awareness were underexplored, despite their importance in empathy, resilience, and reflective practice. Responsible decision-making was the least studied competency, despite its importance in ethical reasoning. Social and emotional learning was consistently associated with enhanced adaptation, communication, leadership, relationships, and clinical performance. Effective strategies included blended learning, simulation, reflective activities, and mentorship, which are aligned with Generation Z's learning preferences. CONCLUSION: Although social and emotional learning integration is associated with improvements in educational and clinical outcomes in nursing, current research has largely centered on relational and stress-related competencies while underrepresenting responsible decision-making. To cultivate reflective, empathetic, and ethically grounded nurses, curricula should integrate social and emotional learning through a balanced and structured approach. REGISTRATION: This study was registered on PROSPERO (ID: CRD420251005683).

Humans↗

PubMed2026

Measuring economic efficiency in adult intensive care units: A systematic review of methods, metrics, and evidence.

OBJECTIVES: Intensive care units (ICUs) consume substantial hospital resources, yet "efficiency" is inconsistently defined and measured. This study systematically reviewed how economic efficiency has been conceptualised and quantified in adult ICUs and appraised the quality of evidence. METHODS: Following PRISMA 2020 and a PROSPERO-registered protocol (CRD420251107866), we searched MEDLINE, Embase, CINAHL, Cochrane Library and Web of Science (2000-August 2025), plus global grey sources. Eligible studies explicitly defined efficiency and reported an efficiency metric/model linking ICU inputs (e.g., staff, beds/capacity, time, consumables, or costs) to outputs/outcomes (e.g., throughput/discharges, length of stay/resource use, risk-adjusted mortality). Dual independent screening and extraction were performed. Study quality was appraised using MMAT, and findings were synthesised narratively (SWiM), given heterogeneity. RESULTS: 39 studies (2001-2025) from 17 countries were included, all from high-income or upper-middle-income settings. Four methodological families were identified: (1) frontier modelling (predominantly DEA; occasional SFA/RFDH), (2) benchmarking indicators (risk-adjusted mortality and LOS/resource-use ratios; "efficiency matrix" quadrant classification), (3) cost-outcome evaluations, and (4) operational/process metrics. Across families, variation in decision-making units, input/output selection, and risk adjustment limited comparability; long-term and patient-reported outcomes were absent, and equity considerations were uncommon. CONCLUSIONS: ICU efficiency research is feasible but fragmented and often methodologically limited. Standardised definitions, validated risk adjustment, uncertainty quantification, and inclusion of patient-centred and equity-relevant outcomes are needed before efficiency metrics can reliably inform value-based decision making.

Intensive Care Units↗

PubMed2026

The role of the external genitalia score (EGS) in evaluation of disorders of sex development.

OBJECTIVE: To investigate the utility of the External Genitalia Score (EGS) in the diagnosis of disorders of sex development (DSD) and decision-making regarding gender assignment in affected patients. METHODS: A retrospective cohort study was conducted, enrolling 114 DSD patients aged <2 years (88 reared as males, 26 reared as females) treated at our hospital between April 2005 and June 2023, alongside 40 hypospadias patients aged <2 years who underwent surgery at our institution from January to July 2023. Demographic data (age) and EGS assessments of external genitalia were collected for all participants. Statistical analyses included independent samples t-tests, Mann-Whitney U tests and Receiver Operating Characteristic (ROC) curve analysis. Specifically, EGS scores were compared between the hypospadias group and the male-reared subgroup of the DSD cohort; additionally, EGS scores were contrasted between male-reared and female-reared DSD subgroups. RESULTS: The mean age was 20.3 months in the hypospadias group, 17.9 months in the male-reared DSD group, and 18.8 months in the female-reared DSD group. EGS ranged from 5.5 to 11.5 (median 10.5) in the hypospadias group and from 1 to 12 (median 4.75) in the DSD group. ROC curve analysis was performed to compare EGS scores between the hypospadias group and the male-reared DSD subgroup. The optimal diagnostic threshold was determined by maximizing the Youden index (sensitivity + specificity - 1), which balances sensitivity and specificity. A cut-off value of ≤8.50 was identified as indicative of DSD; clinically, patients with an EGS score <9 should be prioritized for DSD screening. Further comparison between male-reared and female-reared DSD subgroups yielded a threshold of 4.00. Clinically, an EGS score ≤4 may suggest a preference for female gender assignment. DISCUSSION: The EGS scale is a reliable, valid, and clinically feasible tool for characterizing external genitalia in DSD patients. An EGS score of 9 can serve as an indicator for initiating detailed sex development evaluation in hypospadias patients. While gender assignment in DSD is a complex, multifactorial process, EGS scores showed a significant association with the sex of rearing in our cohort. In settings where major determinants are balanced, EGS may serve as an adjunctive descriptive parameter rather than a standalone decision-making tool.

Humans↗

PubMed2026

Beyond predictive performance: A systematic review and critical methodological appraisal of AI/ML and conventional modelling strategies in breast, colorectal, and pancreatic Cancer.

BACKGROUND: Predictive modelling for cancer risk, treatment-related complications, and survival is central to precision oncology. Conventional logistic regression (LR) and Cox proportional hazards (CoxPH) regression remain widely used but are limited when modelling nonlinear interactions, high-dimensional imaging features, and multimodal clinical-metabolic predictors. Artificial intelligence (AI) and machine learning (ML) methods offer expanded capability through automated feature extraction, ensemble learning, and flexible survival modelling, but the evidence on when AI/ML adds value over conventional models across cancer sites and predictive tasks remains fragmented. OBJECTIVE: To systematically evaluate the methodological performance, validation strategies, and translational limitations of AI/ML models compared with conventional statistical models in published predictive-modelling studies for breast, colorectal, or pancreatic cancer. METHODS: PubMed, Scopus, and Web of Science were searched for studies published between January 2019 and March 2025. Two reviewers independently conducted title-and-abstract screening, full-text eligibility assessment, and PROBAST risk-of-bias assessment. Sixty-five studies (n = 907,567 participants) were narratively synthesised by cancer site, predictive task, model family, comparator, validation strategy, predictor modality, and calibration or explainability reporting. RESULTS: The 65 studies comprised breast cancer (n = 35), colorectal cancer (n = 21), and pancreatic cancer (n = 9). AI/ML superiority over LR and CoxPH was task- and data-dependent. CNN- and U-Net-based models predominated in imaging and body-composition tasks, tree-based ensembles consistently outperformed LR for tabular perioperative complication prediction, and CoxPH remained competitive, and in the largest pancreatic risk study, superior to XGBoost (C-index 0.802 vs 0.723) in well-structured datasets. PROBAST analysis-domain risk was moderate in 54 of 65 studies (83%), driven by limited external validation, sparse calibration reporting (11/65), and few decision-curve analyses (7/65). CONCLUSION: AI/ML adds the most methodological value in imaging-derived feature extraction and nonlinear perioperative prediction, while conventional regression remains preferable in large, structured datasets with linear predictors. Clinical translation requires standardised body-composition definitions, external validation, calibration assessment, decision-curve analysis, and explainability, in line with TRIPOD+AI and CLAIM standards.

Humans↗

PubMed2026

Risk factors associated with urinary tract infection within 4 days of male rectal cancer surgery in the era of enhanced recovery after surgery (ERAS) programs.

BACKGROUND: Bladder drainage is systematically used in rectal cancer surgery in male patients, even in the era of enhanced recovery after surgery (ERAS). However, little data is available on risk factors for urinary tract infection (UTI). Identifying the risk factors associated with UTI within 4 days of male rectal cancer surgery in an ERAS program could support more individualized decision-making. METHODS: We used data from the GRECCAR 10 randomized clinical trial, a comparison of outcomes of transurethral catheterization (TUC) or suprapubic catheterization (SPC). 240 patients were randomized, 209 retained in the study (TUC n = 99; SPC n = 109). Univariate and multivariate logistic regression post-hoc study analyses were performed to assess association between potential predictive factors and UTI within 30 days after surgery. RESULTS: Out of 208 patients (median age 64.5 years), 19 (9.1%) had UTI, 26 (12.5%) had bacteriuria and 145 (69.7%) had pyuria. Univariate analysis identified age ≥ 65 years (OR = 3.08 [1.07-8.89]; p = 0.038), hypertension (OR = 3.65 [1.23-10.84]; p = 0.020) and ASA score ≥ 3 (OR = 4.15 [1.53-11.2]; p = 0.005) as risk factors for UTI until POD4. Multivariate analysis identified ASA score ≥ 3 with a risk of UTI. CONCLUSION: Regarding male rectal cancer surgery, our study shows that nearly 1 in 10 patients had UTI within 4 days. An ASA score ≥ 3 is an independent risk factor linked to UTI. Identifying this risk factor for UTI is necessary to advise patients, support a tailored decision-making process, and prevent these complications.

Humans↗

PubMed2026

Could the preoperative urethral curve be used to predict immediate urinary continence following Retzius-sparing robot-assisted radical prostatectomy? A retrospective multi-center study.

PURPOSE: Immediate urinary continence (UC) recovery following Retzius-sparing robot-assisted radical prostatectomy (RS-RARP) remains highly variable, highlighting the need for reliable preoperative prediction. We aimed to develop and validate models to identify patients likely to achieve immediate UC recovery following RS-RARP. MATERIALS AND METHODS: A total of 580 prostate cancer patients who underwent RS-RARP from four medical centers were assigned to a training set (n=348), an internal validation set (n=103) and an external validation set (n=129). Independent predictors were identified through univariate analysis and LASSO regression. A nomogram was constructed using multivariate logistic regression. Its performance was evaluated with receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis. RESULTS: Immediate UC recovery was observed in 84.5% (294/348) of patients in the training cohort, 80.6% (83/103) in the internal validation cohort, and 81.4% (105/129) in the external validation cohort, respectively. Multivariate analysis identified membranous urethral length (MUL) (OR=1.23, P=0.029) and urethral curvature (OR=2.84, P<0.001) as independent predictors, while prostate volume (PV) (OR=0.84, P <0.001) as a protective factor. The nomogram integrating MUL, PV, and urethral curvature demonstrated superior predictive accuracy, with an AUC of 0.87 (95% CI, 0.83-0.91) in the training cohort. The bootstrap-corrected calibration slope was 0.96, and the Brier score was 0.08. Calibration curves and decision curve analysis confirmed the predictive accuracy and clinical utility of the nomogram. CONCLUSIONS: Our study introduces a novel quantitative method for assessing urethral curvature. The mpMRI-based model, integrating urethral curvature and prostate spatial configuration, offers enhanced predictive accuracy for postoperative immediate UC recovery.

Humans↗

PubMed2026

Complementary feeding patterns in preterm and term infants.

Complementary feeding is essential for infants' nutritional status and development, marking the transition to solid foods when breast milk or formula alone is insufficient. Despite its importance, clear recommendations on which foods to introduce when initiating complementary feeding in preterm infants are lacking. By using data from our previously published randomized controlled trial on the timing of complementary feeding in preterm infants, the current study explores the complementary feeding patterns of preterm infants and compares them with those of term-born infants, providing insights into parental decision-making and potential long-term health impacts. Complementary feeding practices differed significantly between preterm (n = 255) and term (n = 159) infants, with preterm infants more often receiving vegetables as their first solid food (85.4% versus 68.8%, difference 17.6% with 95% CI 12-35%). The group with early introduction of vegetables had a lower BMI-for-age z-scores (β -0.28 [95% CI -0.55 - 0.02]) and weight-for-height z-scores (β -0.27 [95% CI -0.53 to -0.01]) at two years of age. Additionally, preterm infants showed a greater variety in the numbers of different fruits and vegetables consumed by six months (corrected) age than term-born counterparts (8.29 (SD 3.65) versus 6.26 (SD 3.47), p < 0.001). These results indicate that complementary feeding patterns in preterm infants differ from term-born infants, with potential positive implications on growth. These data contribute to the development of accurate feeding protocols for preterm infants. Given that feeding practices are culturally influenced, further multinational research is essential to refine complementary feeding guidelines for preterm infants and support caregivers in informed decision-making.

Humans↗

PubMed2026

A multi-scale fusion model based on multi-phase contrast-enhanced CT for predicting pancreatic cancer resectability.

Purpose.Develop a multi-scale fusion model (MSFM) based on multi-phase contrast-enhanced computed tomography (CECT) to predict pancreatic cancer (PC) resectability, thereby assisting expert decision-making.Methods.This retrospective study enrolled 280 patients with PC from four institutions, which were randomly divided into a training cohort (202 patients) and an independent test cohort (78 patients). Three-phase CECT images (arterial, venous, and delayed phases) were used for modeling. The MSFM comprises two sub-networks: (1) a multi-phase fusion network for extracting cross-phase shared fusion features, (2) a phase-specific branch network for capturing phase-specific features; and a post-fusion strategy to generate the final predictive score by integrating the shared fusion features and three groups of phase-specific features. Additionally, a human-machine fusion deep learning model (HMfDL) was constructed by fusing the predictive score of the MSFM with expert assessments.Results.In the independent test, the MSFM achieved an AUC (area under the receiver operating characteristic curve) of 0.8385 (95% CI: 0.7521-0.9249), accuracy of 84.62%, sensitivity of 72.00%, and specificity of 90.57%. This performance outperformed single-phase models (AUC range: 0.7638-0.7781), two-phase models (AUC range: 0.7826-0.7864), and ten states-of-the-art classifiers (AUC range: 0.7404-0.7796). The HMfDL further improved the performance, reaching an AUC of 0.8626 (95% CI: 0.7853-0.9400), accuracy of 91.03%, sensitivity of 80.00%, and specificity of 96.23%. Notably, the HMfDL corrected 58.82% of misdiagnosis made by experts.Conclusions. The MSFM effectively fuses multi-phase CECT to enable highly accurate predictions of PC resectability, and provides valuable support for expert decision-making through HMfDL.

Humans↗