SEARCH · Search PubMed

Results for “functional magnetic resonance imaging”

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.

954 records · Page 10Linked to original sources

PubMed2026

Physician-Modified Fenestrated Stent-Grafts Planned Using Three-Dimensional Techniques for Complex Aortic Pathology: A Systematic Review and Meta-Analysis.

BACKGROUND: Complex aortic pathology involving the visceral arteries remains a significant therapeutic challenge. Open repair is associated with considerable perioperative risk, particularly in patients with multiple comorbidities, while standard endovascular aneurysm repair (EVAR) is often not feasible because of inadequate proximal sealing zones. Fenestrated and branched endovascular repair (F/BEVAR) represents an established treatment strategy; however, the use of custom-made devices is limited by manufacturing time and availability. Physician-modified stent grafts (PMSGs) have therefore emerged as a pragmatic alternative. Three-dimensional planning techniques have been increasingly used to facilitate accurate graft modification. The aim of this systematic review and meta-analysis was to evaluate the effectiveness and safety of PMSG procedures planned with three-dimensional techniques. Technical success, target vessel patency, early mortality, endoleak occurrence, and reintervention rates were analyzed. METHODS: A systematic search was conducted in the PubMed/MEDLINE and Embase databases. Studies describing the use of physician-modified fenestrated stent grafts planned with three-dimensional tools were included. Meta-analyses were performed using a random-effects model with restricted maximum likelihood estimation. A logit transformation was used for the analysis of proportions. RESULTS: The analysis included five studies involving 172 patients. The estimated weighted mean follow-up duration was 14.9 months. The overall technical success rate was 92.9% (95% confidence interval [CI]: 84.5-96.9%), with low-to-moderate heterogeneity. Target vessel patency was 96.9% (95% CI: 93.6-98.5%). Early mortality was 5.5% (95% CI: 2.1-13.3%). The incidence of endoleaks was 13.3% (95% CI: 5.8-27.4%), with significant heterogeneity among studies. Reinterventions were reported in 6.6% of patients (95% CI: 2.3-17.5%). CONCLUSION: The results indicate that PMSG procedures planned with three-dimensional techniques are associated with a high rate of technical success and preserved patency of target vessels in patients with complex aortic pathology. The observed variability in endoleak and reintervention rates likely reflects differences in anatomical complexity and patient selection among studies. Further prospective studies are needed to confirm long-term outcomes.

Humans↗

PubMed2026

Efficacy of current approaches to non-invasive diagnosis of skin cancer and the potential impact of artificial intelligence: A systematic review and meta-analysis.

BACKGROUND: Skin cancer is one of the most prevalent malignancies worldwide, particularly within Caucasian populations. This systematic review and meta-analysis aimed to quantitatively review the current literature on non-invasive diagnosis of skin cancer and evaluate the current evidence to support the use of tools in addition to, or in replacement of clinician face-to-face assessment. METHODS: A literature search was conducted for publications in PubMed, Medline and Embase databases. Articles describing accuracy, sensitivity, specificity and outcomes of their mode of assessment were included. A total of 208 articles met the inclusion criteria. RESULTS AND CONCLUSION: This systematic review and meta-analysis showed that the diagnostic performance of artificial intelligence (AI) in the interpretation of dermatoscopic images was high for melanoma diagnosis, basal cell carcinoma or malignancy, in comparison to dermatoscopic assessment alone by clinicians and experts. Although AI interpretation of images demonstrated higher sensitivity for melanoma diagnosis in comparison to clinical assessment combined with dermatoscopic assessment, it is unclear if this is also the case for basal cell carcinoma and squamous cell carcinoma diagnosis. Reflectance confocal microscopy, a non-invasive high resolution imaging technique, is known to have a high sensitivity for diagnosing cutaneous malignancy, and this may have applications within secondary care. Therefore, AI could help reduce resource burden and aid in clinical assessment, particularly within primary care settings.

Humans↗

PubMed2026

Depression and amyloid-β across CSF, PET, and plasma biomarkers: a systematic review and meta-analysis.

Alzheimer's disease is increasingly defined by biomarker evidence of amyloid-β and tau pathology, sharpening questions about whether late-life depression contributes to, or instead reflects, this pathology. We conducted a systematic review and meta-analysis of studies published between 2000 and 2025 that compared amyloid-β biomarkers in adults with and without depression, with depression defined by validated clinical diagnoses or symptom rating scales. Twenty-four studies were included, spanning three biomarker sources: cerebrospinal fluid, positron emission tomography imaging, and plasma. Across all sources, the pooled difference in amyloid-β burden between depressed and non-depressed individuals was small and clustered near zero, indicating only a weak, statistically non-significant tendency toward higher amyloid in depression. When the three sources were examined separately, each yielded a similar near-null result, although between-study heterogeneity was considerable for cerebrospinal fluid and plasma and moderate for imaging. Importantly, a prespecified subgroup analysis showed that imaging results diverged by quantification method: studies using the simpler standardized uptake value ratio clustered around zero, whereas the smaller group of studies using kinetic distribution volume ratio modelling showed a significant positive association, suggesting that methodological choices critically influence the observed relationship. Taken together, these findings indicate that depression is not consistently accompanied by greater amyloid-β burden across widely used biomarker platforms. The distribution volume ratio signal nonetheless raises the possibility of subtle associations that cruder methods may obscure, and suggests that depression may shape Alzheimer's disease trajectories more by modifying the clinical impact of amyloid than by altering its amount.

Humans↗

PubMed2026

Artificial intelligence-supported double reading in European population breast cancer screening: A systematic review and meta-analysis of prospective programs.

BACKGROUND: Most European population mammography screening programs rely on double reading with arbitration, a model that delivers mortality benefit but is increasingly challenged by radiologist workload, variable specificity, and interval cancers. Artificial intelligence (AI) is being evaluated to support or optimize these established European screening pathways. PURPOSE: To synthesize prospective or program-embedded evaluations of AI conducted within European-style population screening programs and to estimate exploratory program-level absolute risk differences (RDs) per 1000 examinations for cancer detection rate (CDR) and recall. MATERIALS AND METHODS: We performed a prespecified, focused evidence synthesis of three large studies embedded within routine population screening programs operating under European-relevant workflows: MASAI (randomized AI-supported risk triage within a national program), ScreenTrustCAD (prospective paired-reader evaluation with AI as an independent reader in a double-reading framework), and PRAIM (nationwide decision-referral implementation). Outcomes were harmonized as AI-control RDs per 1000 examinations. Random-effects pooling used Hartung-Knapp-Sidik-Jonkman models. For the paired-reader design, sensitivity analyses applied a Kish effective sample-size approach across plausible within-examination correlations (ρ = 0.3-0.8). Positive predictive value (PPV) and workflow/time outcomes were summarized descriptively. RESULTS: Across 597,419 examinations, the pooled CDR RD was +0.9 per 1000 (95% CI -0.0 to +1.8; I2 ≈ 12%), consistent with a modest directional increase with borderline statistical uncertainty. The pooled recall RD was -0.6 per 1000 (95% CI -3.1 to +2.1; I2 ≈ 41-43%), indicating no consistent recall increase across screening programs. Where reported, PPV was higher with AI-supported screening. Efficiency signals included 44.3% fewer total readings in MASAI and shorter reading times for AI-normal examinations in PRAIM; in PRAIM, a program-level safety-net mechanism recovered 204 cancers that would otherwise have been missed. CONCLUSION: In European population screening programs characterized by double reading and arbitration, prospective program-embedded evidence suggests that AI integration may yield a small absolute increase in cancer detection (≈1/1000) without a consistent increase in recall, alongside improved PPV and efficiency signals. These findings suggestAI primarily as a complementary reader within European screening workflows, with implementation requiring explicit quality assurance and monitoring of interval cancers and stage distribution.

Humans↗

PubMed2026

Granger connectivity and graph-theoretical analysis of scalp EEG across the preictal to ictal transition for presurgical evaluation.

OBJECTIVE: To assess the feasibility of estimating lateralization and localization of the epileptogenic zone (EZ) in temporal and extratemporal lobe epilepsy by combining Electric Source Imaging (ESI) with functional connectivity analysis of high-density EEG from the preictal to the ictal phase. METHODS: Adults with drug-resistant focal epilepsy and at least one recorded seizure during 40- or 64 channels EEG monitoring were retrospectively included. Granger causality and hubness centrality were computed over the 10-s preictal interval and the first 5 s of the ictal period, with ictal onset defined as the first EEG change identified by experienced epileptologists. The reference standard for EZ localization was based on resective surgical outcome or stereo-EEG findings. RESULTS: Thirteen patients (7 females; median age 35 years) were included. Connectivity analyses showed higher concordance with clinical findings during the preictal phase than during the ictal phase for both lateralization (91% vs 46%) and localization (73% vs 27%). Performance was highest in temporal (7/7 lateralization; 6/7 localization) and frontal lobe epilepsy (2/2 for both), and lower in parieto-occipital epilepsy (1/2 and 0/2, respectively). In two cases with poor surgical outcome or no surgical indication, connectivity findings were discordant with clinical estimates. CONCLUSIONS: Connectivity analysis across the preictal to ictal transition provides relevant lateralizing and localizing information, particularly in temporal and frontal lobe epilepsy, and may reveal clinically meaningful discordance. SIGNIFICANCE: Integrating high-density EEG, ESI, and functional connectivity during the phase preceding the first EEG change may support non-invasive presurgical evaluation.

Humans↗

PubMed2026

IgA Vasculitis with necrotizing arteritis: a multicenter retrospective study from the French Vasculitis Study Group and systematic review of the literature.

IgA vasculitis (IgAV) primarily affects small vessels, but rare cases with necrotizing arteritis (NA) raise questions about overlap with polyarteritis nodosa (PAN). To characterize IgAV with necrotizing arteritis (IgAV-NA) and compare its phenotype with classical IgAV and PAN. We performed a multicenter retrospective study combined with a systematic literature review (1990-2025). Patients fulfilled EULAR/PRINTO/PRES IgAV criteria, had pathological or imaging evidence of NA in small or medium arteries, and were ANCA-negative. Thirty patients were included (7 from databases, 23 from the literature). NA was confirmed by biopsy (n = 16) or vascular imaging (n = 14). Clinical features, treatments, remission, and mortality were compared with 257 adult IgAV and 196 PAN patients. Median age was 54.5 years. IgAV-NA was characterized by severe manifestations, including gastrointestinal bleeding, perforation, surgical abdomen, neuropathy, pancreatitis, and livedo. Compared with classical IgAV, IgAV-NA showed significantly higher rates of multi-organ involvement and mortality. Compared with PAN, IgAV-NA shared vascular complications but had less fever and neuropathy. Despite arterial involvement, patients did not fulfil PAN criteria. IgAV-NA represents a rare, severe IgAV phenotype with life-threatening complications rather than an IgAV-PAN overlap. Severe or atypical IgAV presentations should prompt vascular imaging and intensified immunosuppression.

Humans↗

PubMed2026

The Role of Artificial Intelligence Combined With Digital Cholangioscopy for Indeterminant and Malignant Biliary Strictures: A Systematic Review and Meta-analysis.

BACKGROUND: Current endoscopic retrograde cholangiopancreatography (ERCP) and cholangioscopic-based diagnostic sampling for indeterminant biliary strictures remain suboptimal. Artificial intelligence (AI)-based algorithms by means of computer vision in machine learning have been applied to cholangioscopy in an effort to improve diagnostic yield. The aim of this study was to perform a systematic review and meta-analysis to evaluate the diagnostic performance of AI-based diagnostic performance of AI-associated cholangioscopic diagnosis of indeterminant or malignant biliary strictures. METHODS: Individualized searches were developed in accordance with PRISMA and MOOSE guidelines, and meta-analysis according to Cochrane Diagnostic Test Accuracy working group methodology. A bivariate model was used to compute pooled sensitivity and specificity, likelihood ratio, diagnostic odds ratio, and summary receiver operating characteristics curve (SROC). RESULTS: Five studies (n=675 lesions; a total of 2,685,674 cholangioscopic images) were included. All but one study analyzed a deep learning AI-based system using a convoluted neural network (CNN) with an average image processing speed of 30 to 60 frames per second. The pooled sensitivity and specificity were 95% (95% CI: 85-98) and 88% (95% CI: 76-94), with a diagnostic accuracy (SROC) of 97% (95% CI: 95-98). Sensitivity analysis of CNN studies (4 studies, 538 patients) demonstrated a pooled sensitivity, specificity, and accuracy (SROC) of 95% (95% CI: 82-99), 88% (95% CI: 72-95), and 97% (95% CI: 95-98), respectively. CONCLUSIONS: Artificial intelligence-based machine learning of cholangioscopy images appears to be a promising modality for the diagnosis of indeterminant and malignant biliary strictures.

Humans↗

PubMed2026

Robotic-assisted transbronchial biopsy versus computed tomography-guided transthoracic needle biopsy for peripheral pulmonary lesions: a systematic review and meta-analysis of direct comparative studies.

Robotic-assisted bronchoscopy (RAB) and computed tomography-guided transthoracic biopsy (CTTB) are competing strategies for sampling peripheral pulmonary lesions (PPLs). Whether they differ in yield or safety is uncertain. To our knowledge, this is the first systematic review restricted to direct comparisons. We searched MEDLINE, Europe PMC, Scopus, Web of Science and ClinicalTrials.gov from inception to 7 July 2026 for studies directly comparing RAB with CTTB in adults with PPLs. The primary outcome was strict 2024 American Thoracic Society/American College of Chest Physicians diagnostic yield. Risk of bias was assessed with ROBINS-I and certainty with GRADE. A cohort-genealogy step identified, per outcome, the largest set of cohorts sharing no patients; only that set was pooled, with Hartung-Knapp and Mantel-Haenszel sensitivity analyses. Five retrospective studies from one US health system were eligible. Four share patients; at most three cohorts are mutually independent. Across those three, diagnostic yield was comparable (risk ratio [RR] 0.99, 95% confidence interval [CI] 0.93-1.06; I²=24%; Hartung-Knapp 0.87-1.13), with an identical relative effect under strict and intermediate definitions although absolute yields fell from 88% to 74-84% under strict criteria. Pneumothorax requiring a chest tube and/or admission was about three-quarters less frequent with RAB across all three cohorts (RR 0.25, 95% CI 0.14-0.46; I²=0%; Hartung-Knapp 0.07-0.96). Strict yield (RR 0.99) and any pneumothorax (RR 0.06) were reported by two cohorts each and neither survives the few-studies correction. RAB took about 50 min longer than CTTB where same-session staging endobronchial ultrasound was counted in the robotic time, but only about 8 min longer than CTTB where it was not. Only one cohort reported yield by lesion size category and none reported yield by bronchus sign or lung zone, so lesion-level subgroups could not be pooled. Certainty was low for pleural complications and very low elsewhere. Low-certainty evidence indicates that RAB is associated with fewer pleural complications, with no statistically detectable difference in diagnostic yield; equivalence was not formally established. Because all evidence is retrospective, confined to one health system, and almost never stratified by lesion size or accessibility, these findings are hypothesis-generating and require a multicenter randomized trial.

Humans↗

PubMed2026

PdIr bimetallic nanozyme engineered metal-organic frameworks integrated dual-mode sensor toward Stx2 detection in food.

Shiga toxin II (Stx2) has attracted extensive attention due to its toxicity and pathogenicity, making the development of sensitive detection methods urgent. This study constructed a dual-mode sensing platform for the sensitive detection of Stx2 in food. Composite material UIO-66@PdIr with peroxidase-like activity and fluorescent properties was synthesized and combined with cDNA as the signal probe, while aptamer-modified magnetic beads served as the capture probe. Specific binding of Stx2 to the aptamer triggered the release of the signal probe, enabling colorimetric and fluorescence signal readout. The colorimetric mode showed a linear range of 0.05-100 ng/mL with an LOD of 0.039 ng/mL, and the fluorescence mode exhibited 0.01-1000 ng/mL with an LOD of 0.0097 ng/mL. Additionally, this method was successfully applied to the detection of Stx2 in food, and the recovery rates were 94.33% ∼ 102.20%. It indicated that the constructed sensor holds great practical potential for Stx2 detection.

Food Contamination↗

PubMed2026

An Overview of Red Breast Syndrome: A Qualitative Systematic Review of the Literature and a Single-Center Case Series.

Red breast syndrome (RBS) is an uncommon inflammatory complication that may occur following implant-based breast reconstruction and can clinically mimic surgical-site infection. This qualitative systematic review, based on comprehensive PubMed and Scopus searches, summarizes the existing literature on RBS, focusing on its etiology, clinical presentation, diagnostic characteristics, and therapeutic approaches following breast reconstruction using acellular dermal matrices (ADMs). A management protocol is proposed based on available data, and a single-center case series is presented from our Hospital, including patients who developed RBS after direct-to-implant reconstruction with ADM between January 2022 and December 2024. Twenty-nine studies met the inclusion criteria, comprising case reports, case series, and retrospective or prospective cohort studies. The reported incidence of RBS ranged from 0% to 29.6%. Proposed etiologies include endotoxin contamination, residual cellular debris, delayed hypersensitivity reactions, and lymphatic or vascular impairment of mastectomy flaps. Most reports described localized erythema confined to the area of the ADM, with normal inflammatory markers and negative imaging findings, whereas antibiotic therapy frequently failed to achieve improvement. Corticosteroids represented the most consistently effective treatment, although some cases required ADM removal or replacement. In our institutional case series (n = 8), symptom onset occurred between 4 weeks and 7 months postsurgery. Inflammatory markers remained within normal limits, imaging rarely demonstrated fluid collections, and symptoms resolved within days to weeks with corticosteroid therapy or conservative management. This review consolidates current evidence, proposes a diagnostic-therapeutic algorithm and highlights the need for prospective investigations and manufacturer-level endotoxin testing to elucidate pathogenesis and optimize clinical management. Level of Evidence: 2 (Risk) For image description, please refer to the figure legend and surrounding text.

Humans↗

PubMed2026

Multimodal alignment improves generalizability of genomic biomarker prediction in computational pathology.

Computational pathology models that use digitized histopathology whole-slide images have the potential to become a cost-effective and scalable alternative to molecular assays for the prediction of genomic biomarkers, a key task in precision oncology. However, as new genomic biomarkers are discovered or quantified, large, labeled datasets must be prospectively collected to train new models. To address this challenge, we developed multimodal alignment for biomarker learning and generalization (MARBLE), a multimodal contrastive pretraining strategy that integrates structured biomarker knowledge into representation learning of histopathology images. MARBLE aligns histopathology-derived representations with representations of genomic biomarkers generated by a large language model (LLM) and a protein language model (PLM). This biologically informed alignment enables data-efficient generalization to novel, out-of-distribution biomarkers. Using the MSK-IMPACT cohort of over 40,000 patients across multiple biomarker panel versions, we design experiments grounded in real-world data to demonstrate the value of our proposed approach.

CP: computational biology↗

PubMed2026

Development and Crossover Evaluation of an Artificial Intelligence-Assisted System for Solid Pancreatic Lesion Detection and Pancreatic Parenchyma Recognition in Endoscopic Ultrasonography (With Video).

BACKGROUND AND STUDY AIMS: Pancreatobiliary endoscopic ultrasonography (EUS) is technically demanding, and supervised training opportunities are limited. We developed an artificial intelligence (AI) overlay system for detecting solid pancreatic lesions (SPL) and recognizing pancreatic parenchyma (PP) and evaluated its effect on reader performance. PATIENTS AND METHODS: Across six centers, two deep learning-based models were trained using expert-annotated EUS frames. We then conducted a randomized, two-sequence, two-period crossover reader study in which eight endosonographers (five novices and three experts) interpreted image sets with and without AI assistance. The primary endpoint was superiority of sensitivity for SPL detection among novices; key secondary endpoints included specificity and PP recognition. RESULTS: From 118 patients, 120 SPL-positive/negative image sets and 160 PP-positive/negative image sets were constructed. Among novices, AI assistance improved SPL detection sensitivity (88.7% vs. 76.8%, p < 0.001) and accuracy (86.4% vs. 78.7%), while specificity met the predefined noninferiority criterion (84.2% vs. 80.5%, p < 0.001). For PP recognition, sensitivity increased numerically (86.3% vs. 83.3%) but did not meet the predefined superiority criterion (p = 0.095); specificity met the noninferiority criterion (87.8% vs. 81.0%), and accuracy increased from 82.1% to 87.0%. Among experts, sensitivity was maintained for both tasks, whereas specificity increased with AI assistance. CONCLUSIONS: AI assistance improved SPL detection among novice endosonographers. For PP recognition, sensitivity increased without reaching statistical superiority, whereas specificity met the predefined noninferiority criterion. These findings support a potential adjunctive role for AI in EUS interpretation.

Humans↗

PubMed2026

Longitudinal Prediction of Retinal Sensitivity Based on Disease Progression Quantified From Optical Coherence Tomography in Geographic Atrophy.

PURPOSE: The purpose of this study was to analyze the association between disease progression of geographic atrophy (GA) from optical coherence tomography (OCT) with retinal sensitivity (RS) in microperimetry (MP) over a 2-year follow-up period. METHODS: This is a longitudinal analysis of the OAKS Phase-III clinical trial. Both study and fellow eyes with GA that underwent imaging with the Spectralis OCT and consecutive MP examination were eligible. Pointwise quantification of ellipsoid zone (EZ) thickness, EZ and retinal pigment epithelium (RPE) loss from OCT volumes was correlated with localized RS. A longitudinal predictive model using a Markov Chain framework was implemented to predict RS change over time based on OCT biomarkers. The modeling of morphological and functional progression was based on the fellow-eye cohort. RESULTS: A total of 39,681 MP points from 406 patients were analyzed. In the fellow eye cohort, baseline (BSL) EZ thickness was positively associated with RS (0.3 decibel [dB]/µm, P < 0.001). Decrease in EZ thickness between visits during follow-up was significantly associated with decrease in RS (0.1 dB / 1 µm change). RS was significantly lower in MP points within EZ loss during follow-up compared with MP points within the retina with measurable EZ (P < 0.001). The largest functional decline was observed within RPE loss, also associated with the highest probability of absolute scotoma (P < 0.001). Morphological progression to EZ and RPE loss was influenced by EZ thickness and the morphology of adjacent MP points (P < 0.001). CONCLUSIONS: Two exploratory endpoints were developed, namely quantification of EZ thickness and loss, and localized RS within high-risk OCT areas. RS decline during follow-up is associated with automatically quantified disease progression in OCT.

Humans↗

PubMed2026

Biliary Cirrhosis in Myhre Syndrome: The First Case Report of Liver Transplantation and a Review of Reported Hepatic Findings.

Myhre syndrome is a rare autosomal-dominant disorder caused by gain-of-function pathogenic variants in SMAD4 and is now recognized as a progressive multisystem fibrotic disease. Although transforming growth factor-β (TGF-β) signaling plays a central role in hepatic fibrogenesis, hepatobiliary involvement in Myhre syndrome has not been systematically evaluated. We report the first case of Myhre syndrome complicated by rapidly progressive biliary cirrhosis requiring liver transplantation in a 15-year-old male with a confirmed SMAD4 p.Ile500Val variant. Following an infectious episode, the patient developed severe cholestasis with imaging and histopathologic findings consistent with fibro-obliterative cholangiopathy, ultimately necessitating living donor liver transplantation. A systematic review of 55 published reports comprising 217 patients with Myhre syndrome revealed that hepatic evaluation was rarely performed and that previously reported liver abnormalities were mild and secondary, most commonly related to right heart dysfunction or metabolic disease, with no prior cases of progressive biliary fibrosis. This case suggests that dysregulated SMAD4-TGF-β signaling may predispose selected organs to fibro-obliterative injury and that infection-driven inflammation may act as a critical trigger for hepatic fibrosis in Myhre syndrome, expanding the recognized spectrum of organ involvement in this disorder.

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

Vestibular schwannoma associated normal pressure hydrocephalus: clinical features and shunt responsiveness compared with idiopathic NPH.

BACKGROUND: Vestibular schwannoma (VS) is commonly associated with obstructive hydrocephalus due to mass effect; however, a rarer communicating form resembling normal pressure hydrocephalus (NPH) has also been described, possibly related to impaired CSF absorption from elevated CSF protein. We aimed to characterize the clinical and imaging features of VS-associated NPH (VS-NPH) and compare them with those of an idiopathic NPH (iNPH) cohort. METHODS: We retrospectively analyzed 18 patients with VS-NPH identified between 2008 and 2024. For comparison, 41 iNPH patients were drawn from a prospective longitudinal study at our center. Variables included demographics, tumor size, VS treatment modality, CSF parameters, Radscale imaging features, and shunt responsiveness. RESULTS: VS-NPH patients had markedly higher CSF protein levels than patients with iNPH (median 100 vs. 51 mg/dL, p < 0.001). Radiological features largely overlapped; however, parasagittal sulcal narrowing was more frequent in VS-NPH (61 % vs.13 %, p = 0.002). These differences remained significant in the sensitivity analysis excluding the two patients without gait impairment. VS-NPH patients were younger in the primary analysis (66.8 vs. 72.0 years, p = 0.03), while exploratory associations between larger tumor size and both earlier NPH symptom onset (r = -0.48, p = 0.049) and smaller callosal angle (r = -0.49, p = 0.048) attenuated to non-significant trends in the sensitivity analysis. Tumor size was < 30 mm in 89 % of patients. Ventriculoperitoneal shunt (VPS) resulted in clinical improvement in both groups, although response rates were numerically lower in VS-NPH than in iNPH (63 % vs.75 %). CTT was positive in 9 of 11 VS-NPH patients who underwent testing, although improvement after shunting also occurred in patients with negative CTT results or without prior CTT. CONCLUSIONS: VS-NPH may represent a secondary subtype of NPH with distinct biochemical and subtle imaging features. Elevated CSF protein may contribute to altered CSF dynamics. These findings are exploratory and require confirmation in larger prospective studies.

Humans↗

PubMed2026

Decoding the spatiotemporal patterns of food spoilage microbial communities: Integrating multi-omics and artificial intelligence to enable precision preservation.

In the global food supply chain, food wastage caused by spoilage has resulted in significant economic losses, food shortages, and environmental pressure. This process is fundamentally driven by the spatiotemporal dynamics of microbial communities. However, traditional research methods struggle to elucidate the complex mechanisms of spatial heterogeneity, interspecies interactions, and functional succession. This limits the development of effective preservation strategies. This review systematically reviews the cutting-edge progress of integrating multi-omics technologies and artificial intelligence (AI) to study food spoilage microbial communities, breaking through this bottleneck. We propose an intelligent theoretical framework that could potentially analyze microbial metabolic activities and predict dynamic shelf life if implemented. The conceptual framework integrates multidimensional data, including spatial metabolomics, temporal metatranscriptomics, single-cell transcriptomics, and longitudinal metagenomics. It can also be combined with AI models, such as graph neural networks. The article elaborates on the principles and applications of spatio-temporal monitoring technologies, such as nano secondary ion mass spectrometry, hyperspectral imaging, and the Internet of Things sensing. Through illustrative cases of typical perishable foods, it also explores how such a multi-omics - AI system might be applied to spoilage warning and precise intervention. Additionally, the article addresses the current challenges in data coverage, model generalization, and federated learning implementation. Then the research further explores emerging areas such as engineered probiotics, edge AI, and microfluidic sensing. These areas are targeted at transforming food preservation from an empirical control approach to a data-driven, precise regulatory framework. This transformation provides theoretical support and technical approaches for developing a smart, sustainable food preservation system.

Multiomics↗

PubMed2026

Beyond Photometric Consistency: Addressing Loss Insensitivity to Depth Noise in Endoscopic Estimation via Error Calibration.

Self-supervised monocular depth estimation in endoscopy is fundamentally constrained by the ill-posed nature of photometric supervision. In this work, we identify a critical yet overlooked cause of this ambiguity: the inherent insensitivity of photometric loss to depth noise. To overcome this intrinsic limitation, we propose Depth Error Calibration Learning (DECL), a two-stage framework that suppresses prediction variance and mitigates residual errors in self-supervised depth estimation. In Stage I (Variance Reduction), a cyclic depth generation strategy produces multiple depth hypotheses for the input image. The per-pixel empirical variance is quantified and integrated into a dedicated variance loss term, which penalizes inconsistent predictions and encourages the network to generate more stable and reliable depth estimates. In Stage II (Bias Calibration), an image-conditioned diffusion model refines the Stage-I depth prior and mitigates structured residuals through iterative denoising, thereby improving geometric accuracy and global consistency. Extensive experiments on three public endoscopic datasets demonstrate that DECL achieves consistent improvements over representative self-supervised monocular depth estimation methods under the evaluated protocols. Moreover, ablation studies on two representative backbones indicate that DECL is not restricted to a single network implementation, while broader validation on additional backbone families remains necessary. The source code is publicly available at https://github.com/DavidLuBit/EndoDenoising.

Journal Article↗