Biology Methods and Protocols

Background. Comparing long-term outcomes of gastric cancer surgery between hospitals requires a common reference. Risk-adjusted benchmarks depend on the model and population used to build them and are difficult for another hospital to reproduce. TNM stage is a prognostic classification already recorded for every resected gastric cancer. We report unadjusted stage-specific overall survival from a consecutive cohort with complete follow-up as reference data and propose a minimal dataset for producing comparable figures elsewhere. Methods. The cohort comprised 690 consecutive eligible patients who underwent curative-intent D2 gastrectomy for stage I to III adenocarcinoma at Ruijin Hospital, Shanghai, in 2019. All operations were performed by eight experienced, high-volume consultant surgeons. Stage followed the AJCC 8th edition: pTNM after upfront surgery (n = 570) and ypTNM after neoadjuvant chemotherapy (n = 120). Overall survival was estimated by the Kaplan-Meier method without adjustment. Results. There were 199 deaths, 181 within five years. Every survivor was followed for at least 68.1 months. After upfront surgery, five-year overall survival was 95.0% (95% CI 91.4 to 97.1) for stage I, 83.8% (76.5 to 89.0) for stage II and 56.5% (49.2 to 63.1) for stage III, and was 95.0% in IA and 95.2% in IB, falling to 26.7% in IIIC. After neoadjuvant chemotherapy, five-year survival was 78.6%, 61.9% and 31.3% for ypStage I, II and III. Pooling pTNM and ypTNM lowered stage III five-year survival to 50.2%. The proposed minimal dataset has 16 items, 13 of them recorded during the admission for gastrectomy. Conclusions. These figures are intended as a reference for comparing hospitals rather than as a standard of care. We propose 16 items, most of them routinely recorded, from which another hospital can produce the same table and compare its results by stage.

BackgroundEvaluating fairness in artificial intelligence (AI) is essential before clinical implementation to prevent unintended harm across diverse patient populations. This study assesses performance differences of an image-based surgical site infection (SSI) detection model across age, gender, geography, race, and skin color. MethodsModel development/validation study using retrospective, multicenter data across surgical subspecialties from adults who submitted an image of their incision via the electronic health record within 30 days of their procedure between 2019-2024. Area under receiver operating curve (AUROC) was compared by age (<55 vs. [&ge;]55 years old), gender (male vs. female), geography (Rochester, Minnesota vs. Jacksonville, Florida vs. Scottsdale, Arizona vs. regional Minnesota/Wisconsin areas), race (White/Caucasian vs. non-White/Caucasian), and skin color (Fitpatrick I-III vs. IV-VI and Monk A-E vs. F-J). ResultsTraining data contained 13,702 images from 4,274 patients between 2019-2022 (50.1% [&ge;]55 years old; 63.2% female; 91.6% White/Caucasian; 58.5% Rochester, 17.3% Florida, 10.5% Arizona; 96.3% light skin tone), of which 1,348 images (9.8%) had an SSI. No differences in SSI detection existed for age, gender, race, or skin color (AUROC 0.83-0.87, p>0.09), however geographical differences were significant: lowest AUROC 0.83 (95% CI 0.77-0.88) vs. highest AUROC 0.87 (95% CI 0.84-0.89, p=0.01). Temporal validation using 4,242 images from 1,353 patients between 2022-2023 showed intact fairness across subgroups ({Delta}AUROC [&le;]0.06) except geography ({Delta}AUROC [&le;]0.16). ConclusionsThe model behaved fairly across age, gender, race, and skin color, however geographical differences persisted. These findings have practical implications underscoring the importance of evaluating bias in clinical AI before deployment.

Accurate classification of lung adenocarcinoma growth patterns from hematoxylin and eosin (H\&E)-stained whole-slide images is essential for treatment planning and prognosis. We develop and evaluate an exponentially weighted ensemble deep learning approach that combines five architectures (EfficientNet-B3, DeiT3, Swin Transformer, Vision Transformer, ConvNeXt) with weights determined by a softmax transformation of per-model Quadratic Weighted Kappa (QWK) scores. Using 25,545 image patches at 40x magnification from the Dartmouth-Hitchcock dataset, we evaluated performance with stratified 5-fold cross-validation. Individual model performance ranged from Cohen's kappa kappa = 0.5563 (ConvNeXt) to kappa = 0.9561 (EfficientNet-B3). The ensemble achieved 96.52% accuracy and kappa = 0.9648, exceeding the best individual model by delta kappa = +0.0087 with statistical significance (McNemar's test, chi^2 = 50.84, p < 10^-12; bootstrap 95% CI on delta kappa [+0.0056, +0.0117]. Pairwise disagreement rates of 16.6--43.2% between architectures indicate that the models capture different aspects of tissue morphology, and the framework automatically down-weights unreliable components, as illustrated by ConvNeXt receiving only 5.7% ensemble weight despite a single-fold training failure. The approach offers a clinically grounded method for ensemble aggregation in digital pathology.

Background. In people with type 2 diabetes (T2D), postprandial glucose weighs on glycaemic control, and carbohydrate counting is difficult. Patients can describe their meal but rarely quantify it. Objective. To assess (1) whether a small open large language model (LLM), fine-tuned on public data, estimates the carbohydrate content of a meal described in words, and (2) what this carbohydrate adds to the prediction of the postprandial glucose rise in two public cohorts. Methods. Step 1: the Qwen3-4B LLM was fine-tuned with QLoRA on Nutrition5k dishes and compared with an automatically matched USDA food composition table and with Qwen3-4B and Qwen3-8B used without training, on test dishes (internal control) and on 100 real meals of ShanghaiT2DM patients, whose reference carbohydrate was computed from the recorded quantities. Step 2: the maximum glucose rise between 0 and 120 minutes ({Delta}Gmax) was predicted by personalised gradient boosting (M2), compared with a spline linear regression and with the Chronos-2 foundation model, using only information available at meal start and person-level cross-validation, in ShanghaiT2DM (people with T2D recruited in two Shanghai hospitals; carbohydrate estimated by the LLM) and CGMacros (free-living adults in Santa Barbara, California; carbohydrate entered in the dataset, no text). Step 3: the model was transferred between datasets. 95% confidence intervals (CI) were obtained by cluster BCa bootstrap (2,000 resamples). Results. On the 100 real meals, the carbohydrate mean absolute error (MAE) of the fine-tuned model was 12.2 [9.9, 15.2] g, almost half that of the table (22.6 g; paired difference -10.5 [-16.8, -6.0] g) and nearly a third of that of the same model without fine-tuning (34.3 g), with a bias of -10.3 g; on test dishes it was 2.2 [2.0, 2.5] g. Personalised gradient boosting (M2) reached a correlation between predicted and observed rises of r = 0.63 [0.55, 0.69] in CGMacros (1,286 meals, 45 adults) and 0.49 [0.42, 0.54] in ShanghaiT2DM (2,339 meals, 99 people with T2D), about twice the linear reference (r = 0.32 and 0.23); for a rise above 50 mg/dL, its area under the ROC curve (AUROC) was 0.80 and 0.74, versus 0.64 and 0.68 with pre-meal glucose alone. Adding carbohydrate to pre-meal glucose changed r by 0.095 [0.068, 0.128] in CGMacros and by 0.004 [-0.010, 0.017] in ShanghaiT2DM; difference between datasets: 0.091 [0.060, 0.126]. Applied from one dataset to the other, the model lost correlation ({Delta}r = -0.11 [-0.17, -0.05] towards ShanghaiT2DM and -0.12 [-0.19, -0.04] towards CGMacros). The full "text [-&gt;] LLM [-&gt;] glucose" chain was evaluated in ShanghaiT2DM only. Conclusion. An open 4-billion-parameter LLM, fine-tuned with QLoRA on public data and usable offline, almost halves the error of carbohydrate counting for real meals described in words compared with an automatically matched table, with an underestimation of large meals. Personalised gradient boosting, based on pre-meal glucose and the person's usual response, doubles the correlation of a linear reference for predicting the glucose rise in both cohorts. Carbohydrate improved this prediction in CGMacros but not in ShanghaiT2DM; the two datasets differ in many respects, and the cause of this gap is not established.

Context Current guidelines for primary aldosteronism (PA) recommend screening all patients with hypertension. Given low screening rates, an automated artificial intelligence (AI)-based risk-stratified approach may facilitate guideline implementation. Objective To develop an AI-based prescreening tool using electronic health record (EHR) data to identify at-risk patients. Design Model development retrospective cohort study. Ensemble architecture using eXtreme Gradient Boosting, random forest, and extremely randomized trees was trained on EHR data (1986-2025) and evaluated on a test set of patients with hypertension. Setting Single institution system spanning community clinics and tertiary/quaternary centers reported via Mayo Clinic Platform. Patients Adults with hypertension, International Classification of Disease (ICD) diagnosis of PA, or had plasma renin/aldosterone testing. Main Outcome Measures Predicting PA diagnosis (disease) or negative workup (control) using age, sex, ICD diagnoses, vitals, labs, and medications by area under receiver operating characteristic (AUROC), estimated calibration error (ECE), and specificity. Results Of 22,264 patients (1,833 disease and 20,431 controls), median time from initial encounter to screening was 5.0 years (IQR 0.5-12.6). The model demonstrated AUROC 0.709 (95% CI 0.688-0.730) with ECE 0.037. At a 0.5 cut-off threshold, the model had high specificity (71.6%) and low false positivity (28.4%). In a test cohort of 1,279,455 adults with hypertension, the model identified 5,295 (0.4%) for first-pass screening. Conclusions A high-specificity model can provide a pre-screening PA pathway using routine EHR data to identify and prioritize patients for PA screening. This AI-based approach can risk-stratify patients, guide sequencing of testing/referrals, and pragmatically facilitate adoption of current guidelines into practice.

Background: Robust evidence linking individual surgeon technical performance to patient outcomes in complex minimally invasive surgery remains scarce. Crowd-sourced video assessment offers a scalable alternative to expert review, but its clinical relevance in complex oncologic procedures such as robot-assisted minimally invasive esophagectomy (RAMIE) is unestablished. Methods: In this exploratory study, video recordings of 60 consecutive RAMIE procedures performed by 7 surgeons (2018-2023) were segmented into five representative surgical steps and rated by 93 blinded crowd-raters per segment using a modified GEARS instrument. Tertiles were used to compare top-performing operations and steps against the remaining two terciles for overall and step-specific performance, and associations with perioperative outcomes were analyzed. Results: Top-tercile overall performance was associated with significantly shorter operative time (abdominal, thoracic, and total) and less intraoperative blood loss (all p<0.05), without differences in postoperative outcomes. In exploratory step-specific analyses, top-tertile performance during infracarinal lymphadenectomy was associated with significantly fewer surgical complications (30.0% vs 60.0%; p=0.028), fewer severe surgical complications (25.0% vs 52.5%; p=0.043), and higher lymph node yield (22.8 vs 18.9; p=0.037), while the purse-string suture step showed the broadest operative-time benefit. These step-specific findings are hypothesis-generating and were not adjusted for multiple testing. Conclusion: Crowd-sourced video assessment enables scalable, step-specific evaluation of surgical performance in RAMIE associated with intraoperative blood loss and operative time. Exploratory analyses suggest that the relationship between technical performance and clinical outcomes might be step-specific, with infracarinal lymphadenectomy emerging as risky step linked to postoperative morbidity. These findings support further investigation of individualized performance assessment for quality improvement and risk stratification.

Background: Machine Learning (ML) has shown promise in histopathological diagnosis, but rare tumours, like duodenal epithelial neoplasms, are challenging because large, labelled datasets are difficult to assemble. We propose a data-efficient ML pipeline for detecting duodenal epithelial neoplasia in whole-slide images. Methods: We assembled a five-hospital dataset of haematoxylin and eosin-stained duodenal biopsy whole slide images (WSIs) at 10x magnification, comprising 8,027 normal and 401 neoplastic biopsies, including adenomas, carcinomas and neuroendocrine tumours. WSIs were divided into patches and encoded using the Hibou-B foundation model. We compared global standardisation with unsupervised site-specific standardisation of patch embeddings. Our pipeline combined principal component analysis, a Gaussian mixture model and a Random Forest classifier. We assessed generalisability using leave-one-hospital-out cross-validation across four hospitals and on an independent test set from a fifth hospital. Results: In leave-one-hospital-out cross-validation, the pipeline achieved a mean accuracy of 90%, sensitivity of 96%, specificity of 83% and an AUC of 0.96. On the external test hospital, it achieved 94% accuracy, 93% sensitivity, 95% specificity, and AUC of 0.97. Site-specific standardisation improved mean cross-validation accuracy by 20 percentage points compared with global standardisation, while the Gaussian mixture pipeline improved accuracy by 12 percentage points compared with mean pooling of patch embeddings. Conclusions: We present a data-efficient ML approach for detecting duodenal epithelial neoplasia from WSIs. Strong performance was maintained across hospitals and on an independent external test cohort despite the limited number of neoplastic training cases, supporting the feasibility of foundation-model-based approaches for rare histopathological diagnoses.

Background: Clinicopathologic (SARCULATOR) and transcriptomic (CINSARC) models are established prognostication tools in soft tissue sarcoma, but neither directly represent the sequence context of tumor genomic alterations. Evo2 is a genomic foundation model trained to learn the grammar of DNA/RNA sequences. We tested whether Evo2-derived features from tumor short variants, fusions, and rearrangements improve survival prognostication beyond clinicopathologic and transcriptomic risk models. Methods: We analysed formalin-fixed, paraffin-embedded tumor material and clinical data from 102 patients with dedifferentiated liposarcoma (n = 25), high-grade leiomyosarcoma (n = 25), and undifferentiated pleomorphic sarcoma (n = 52). A separate TCGA-SARC cohort of comparable histologies (n = 129) served as external validation. Reconstructable alterations (short variants, fusions and rearrangements) were converted into sequence inputs, embedded with Evo2 and aggregated into patient-level genomic representations. Cox proportional hazards models incorporated genomic features alone (Evo2) or together with clinicopathologic (SARCULATOR) or transcriptomic (CINSARC) predictors across overall survival (OS), progression-free survival (PFS), disease-free survival (DFS), relapse-free survival (RFS), and metastasis-free survival (MFS). Discrimination was assessed using Harrell's concordance index (C-index). Results: Evo2 features were generated from 174 study-cohort sequences and 7298 TCGA-SARC sequences. In the study cohort, adding Evo2 to SARCULATOR increased the reported C-index for OS from 0.620 to 0.770, PFS from 0.625 to 0.755, DFS from 0.612 to 0.714, and RFS from 0.633 to 0.771. In TCGA-SARC, SARCULATOR plus Evo2 increased the OS C-index from 0.595 to 0.716 and the MFS C-index from 0.548 to 0.667. Also in TCGA-SARC, adding Evo2 to CINSARC increased the OS C-index from 0.476 to 0.653 and the MFS C-index from 0.553 to 0.650. These are descriptive point-estimate comparisons. Moreover, in TCGA-SARC Evo2 alone outperformed SARCULATOR alone and CINSARC alone, supporting independent prognostic information within altered sequence-context representations. Conclusions: Evo2-derived genomic representations improved clinicopathologic and transcriptomic prognostication across sarcoma cohorts and survival endpoints. These findings support genomic foundation models as an independent and complementary layer for sarcoma risk stratification through direct modeling of altered tumor sequence biology. Confirmation of model transportability, calibration and clinical utility is required before clinical application.

An Ischemic stroke is one of the main causes of death and long-term disability around the world. There are very few treatment options available, especially for the cases where the short time period to give clot dissolving medicine has passed (Powers et al., 2019). While single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool for understanding cell-to-cell differences within at issue (cellular heterogeneity) (Tang et al., 2009). However translating these high dimensional datasets into clinically useful biomarkers remains challenging (Rajko-mar et al., 2019). A major problem with current biomarker studies is their reliance on a single machine learning model, where results may reflect model bias rather than true biological signals (Lopez-Rincon et al., 2020). Methods: We developed a multi-model consensus framework integrating eight different machine learning methods from three groups: traditional models (logistic regression, random forest, XGBoost), a hybrid deep learning model (CNN-XGBoost), and transformer-inspired models (scGPT, scBERT, sc-Former, Geneformer). The proposed framework was applied to scRNA-seq data from a mouse middle cerebral artery occlusion (MCAO) model (GSE174574; n=6 mice, 54,599 cells). To prevent data leakage, we implemented rigorous subject-wise train/test splitting, ensuring all cells from individual animals remained either in the training sets or in the testsets. Results: All eight models achieved excellent performance on the independent test data, with AUC scores between 0.9778 and 0.9975. From six models that allowed gene level importance extraction, we identified 63 consensus genes (threshold: [&ge;] 4/6 models), with 17 genes selected by all 6 models. Pathway enrichment analysis showed significant enrichment of interferon signaling, with a novel dual-interferon pattern. Twelve of the consensus genes respond to interferon, and five of them (Ifitm6, Ifitm1, Gbp2, Igtp, Tgtp1) respond to both Type I (IFN-/{beta}) and Type II (IFN-{gamma}) interferons. Notably the consensus included Il1rn (interleukin-1 receptor antagonist), the target of Anakinra -- an FDA-approved drug that has completed Phase II stroke trials with promising safety profiles (Smith et al., 2018; Emsley et al., 2005). Conclusions: The Multi-model consensus framework, proposed in this paper, provides a robust and reliable approach to biomarker discovery that goes beyond the limitations of any single method and reduces individual algorithm bias. The convergence of classical, ensemble, and deep learning methods on interferon and IL-1 pathway genes supports their biological validity as stroke biomarkers. Our findings provide a prioritized gene list for further clinical validation and highlight interferon signaling as a possible treatment strategy warranting further investigation. Keywords: ischemic stroke; machine learning; single-cell RNA sequencing; biomarker discovery; consensus analysis; interferon signaling; neuroinflammation

Background: The Central Brain Tumor Registry of the United States (CBTRUS) uses a histology grouping model based on the World Health Organization (WHO) to classify cancer registry records for clinically relevant statistical reporting. Methods: CBTRUS, together with its consulting neuropathologists, reviewed and aligned its histology groupings with the 2021 WHO Classification update. Obsolete histology nomenclature along with their International Classification of Diseases for Oncology 3rd edition (ICD-O-3) codes were identified, currently included histologies were reclassified as appropriate, and new codes found in 2021 WHO were incorporated. An evaluation was conducted of the frequency of histology codes affected in the reclassification process, as well as incidence and survival pre- and post- reclassification of CBTRUS Tumor Type Groupings. Results: After review of the CBTRUS grouping scheme, 205 codes were reclassified and 2 new codes were incorporated. Tumor type groups most affected were those classified as glioma. Reclassification resulted in 167,038 of 491,804 (34%) cases having a grouping reclassification or reporting change, indicating that the 2021 revision of the WHO Classification has significantly impacted the collection and reporting of primary b CNS tumors. Conclusion: This work demonstrates the need to be responsive to changes in classification and coding for statistics to reflect real-world clinical use. This will require collaboration from all parties to have the ability to reconcile clinical practices and surveillance requirements.

Objectives: Intraoperative ultrasound (IOUS) is an increasingly valuable real-time imaging modality in neurosurgery, enabling visualization of brain parenchyma, tumor margins, ventricular anatomy, and vascular structures during cranial procedures. Despite its clinical relevance, the development of artificial intelligence (AI)-assisted interpretation tools for neurosurgical IOUS has been limited by the scarcity of large, annotated datasets. Manual pixel-level annotation of ultrasound video frames is labor-intensive, requires specialized neuroanatomical expertise, and scales poorly. To address this bottleneck, we developed a proof-of-concept interactive semi-automated annotation pipeline designed to explore the feasibility of scalable, expert-guided labeling for intraoperative neurosurgical IOUS. Methods: We implemented the system as a napari-based graphical interface coupled with a scikit-learn Random Forest (RF) pixel classifier. Real intraoperative neurosurgical ultrasound videos from five craniotomy cases were used as training material. A neurosurgical expert annotated a minimal set of representative frames using brush-based labeling to define anatomical classes of interest. For each labeled pixel, a feature vector was computed comprising raw intensity, multi-scale Gaussian filtering ({sigma} = 1, 2, 3), Laplacian of Gaussian, and Sobel edge magnitude. The RF classifier was trained on accumulated annotations using class-balanced weighting and out-of-bag (OOB) validation, then applied across full video sequences. An incremental accumulator module enabled multi-case training. Segmentation outputs were exported as TIFF overlays for expert quality review. Results: Across five intraoperative cases, sparse expert annotation of a small number of representative frames per case was sufficient to train case-specific Random Forest classifiers that propagated labels across the full video sequences. Out-of-bag estimation served as an internal coherence check during training, and the classifier additionally produced per-class probability maps that exposed regions of uncertainty at acoustic boundaries. On expert qualitative review, the auto-segmented frames were anatomically coherent, with the classifier reliably distinguishing anechoic fluid compartments from echogenic tissue interfaces; residual misclassification was concentrated in low-contrast boundary regions. Conclusions: This proof-of-concept study demonstrates that semi-automated, expert-guided annotation of intraoperative neurosurgical ultrasound is technically feasible using classical machine learning. By combining neurosurgical domain expertise with an interactive, incremental annotation workflow applied to real intraoperative acquisitions, this framework represents a foundational step toward scalable generation of labeled datasets for neurosurgical IOUS. Future directions include formal multi-rater validation, expanded multi-class anatomical labeling, and eventual migration to deep learning architectures as annotated datasets grow.

High low-density lipoprotein cholesterol (LDL-C) and triglyceride (TG) levels are associated with heavy alcohol consumption, but the accuracy of LDL-C estimation equations according to drinking status remains unclear. We analyzed direct LDL-C and drinking-status data from 28,882 participants in the Korea National Health and Nutrition Examination Survey, classified as non/ex-drinkers, moderate drinkers, or binge drinkers. Mean absolute deviation (MAD) was calculated for the Friedewald, Martin-Hopkins, and Sampson equations relative to directly measured LDL-C, and 95% confidence intervals for the median absolute deviation (MeAD) were determined by TG level. Estimation error increased with heavier alcohol consumption and higher TG level. In binge drinkers, all three equations underestimated LDL-C, with the largest deviation at higher TG levels. The Sampson equation showed the lowest MAD and MeAD across drinking categories and TG strata and was most accurate at a TG level of 150-400 mg/dL, whereas the Martin-Hopkins equation was more accurate at TG levels of [&ge;]400 mg/dL. The Sampson or Martin-Hopkins equation estimates LDL-C more accurately than the Friedewald equation regardless of drinking status and should be preferred, with the choice further individualized by TG level in heavy drinkers.

Purpose: Existing prognostic models for brain metastases from lung cancer rely on clinical variables and do not capture the biological heterogeneity of host or tumor factors underlying survival outcomes. We evaluated whether MRI-derived radiomic biomarkers improve prognostic stratification beyond clinical variables. We also investigated their biological correlates using matched histopathology. Experimental Design: This retrospective study included 103 patients with brain metastases from lung cancer and matched pre-operative MRI and histopathology. Radiomic features from T1-weighted contrast-enhanced and T2-weighted fluid-attenuated inversion recovery sequences were combined with clinical variables to train Cox proportional hazards, random survival forest, and DeepSurv models for overall survival (OS), CNS progression, and leptomeningeal (LM) progression. Risk groups from an imaging-only model were compared for clinical and histopathologic differences. Results: Imaging features improved prognostic performance beyond clinical variables across all modeling frameworks, with the combined clinical-imaging DeepSurv model achieving the highest concordance index for OS (0.728). Risk groups from the best-performing model for each endpoint differed significantly in OS (median 26.7 vs. 9.1 months, p < 0.001) and LM progression (p = 0.003), but not in CNS progression. Imaging-only risk groups did not differ significantly in any clinical variable, but high-risk patients were less likely to have tumor-infiltrating lymphocytes (46% vs. 77%, p = 0.010) and PD-L1 expression [&ge;] 50% (10% vs. 32%, p = 0.049). Conclusions: MRI radiomic biomarkers provide prognostic information beyond established clinical variables in lung cancer brain metastases and correspond to features of the tumor immune microenvironment, supporting their potential to refine risk stratification.

INTRODUCTION: The digital clock drawing test (dCDT) is a cognitive screening tool capturing high-resolution pen-movement data. Existing approaches for classifying cognitive impairment from dCDTs rely on selected features or image-based methods. We propose representing dCDTs as structural networks to classify cognitive impairment and characterize spatial organization in clock drawings. METHODS: Weighted networks were constructed from 3,362 Framingham Heart Study participants, with and without cognitive impairment. Network representations were generated separately for command and copy conditions, with nodes representing pre-defined clock components (digits, hands, clock face) and edges capturing pairwise spatial relationships between components. Random forest models and graph convolutional networks were trained to predict cognitive impairment from the networks. Models were benchmarked against demographics-only and established summary feature baselines. RESULTS: Network-based models achieved AUC values comparable to the established summary feature baseline across command-only and copy-only conditions. Differences in spatial organization between the cognitively intact and impaired groups were greatest in the upper-right clock region, involving digits 1 and 2, the clock hands, and the surrounding clock face. Differences between the command and copy networks were larger among cognitively impaired participants. DISCUSSION: The proposed network representations provide an objective, scalable, and flexible framework for representing clock drawings for use in classification models. dCDT networks matched the predictive performance of established summary feature-based approaches and identified clock regions where spatial organization differed by cognitive status. After external validation, this approach has the potential to be incorporated within a clinical workflow to support automated and interpretable cognitive screening.

Foundation models (FMs) have rapidly become dominant in artificial intelligence and are increasingly being adopted in computational pathology. Numerous pathology-specific FMs have been developed and evaluated for a variety of downstream tasks, most commonly using frozen image embeddings with linear probes. While a pathology FM, at least implicitly suggests broad reusability across tasks, the transferability of these models to neurodegenerative disease-related tasks remains largely unexplored. In this study, we evaluated fourteen frozen feature extractors, including general-vision and pathology FMs, across four pathology image classification datasets spanning two specialized neurodegenerative disease immuno-histochemistry (IHC) tasks (Tau neurofibrillary tangle (NFT) and amyloid-{beta} plaque classification) and two cancer-related hematoxylin and eosin (H&E) tasks (the breast tissue BACH dataset and the multi-tissue TIL dataset). The cancer-related H&E datasets represent tasks more closely aligned with the predominant pretraining domain of many current pathology FMs, whereas the neurodegenerative IHC datasets represent a more specialized domain with limited representation in existing FM pretraining cohorts. Frozen embedding linear probes were compared against a conventional supervised ResNet-50 convolutional neural network (CNN) trained directly on image tiles using identical train, validation, and test splits. Across both neurodegenerative IHC datasets, the supervised CNN substantially outperformed all frozen FM linear probes. In contrast, pathology FMs achieved performance comparable to, and in some cases exceeding, the supervised CNN across the two cancer-related H&E datasets, with UNI2-h achieving the highest performance on BACH and several pathology FMs performing on par with the CNN on the larger TIL dataset. Furthermore, a supervised CNN trained using only 1% of the Tau NFT training data (1,938 tiles) still exceeded the performance of the best frozen FM linear probe trained on the complete dataset. Together, these results suggest the transferability of frozen pathology FMs may depend strongly on how well the downstream task is represented by their pretraining domain. These findings demonstrate frozen pathology FMs transfer effectively to the cancer-related H&E tasks evaluated here but may be less effective for specialized neurodegenerative IHC tasks less well represented in current FM pretraining cohorts. Expanding pathology FM pretraining datasets to include a broader range of disease domains and staining modalities may therefore improve transferability to specialized pathology applications. Our findings also highlight the continued importance of conventional supervised learning and motivate future work investigating nonlinear probes and end-to-end FM fine-tuning.

Objective: To determine whether plasma neurofilament light chain (NfL) and glial fibrillary acidic protein (GFAP) are elevated in hospitalised patients with delirium relative to patients with dementia and primary psychiatric disorders (PPD). Methods and analysis: This was a cross-sectional observational study conducted at the Royal Melbourne Hospital, Melbourne, Australia, between April and October 2024, with comparator samples from a research biobank. In Phase 1 (Delirium Discovery Cohort), we measured plasma NfL and GFAP using the Single molecule array (Simoa) platform in hospitalised patients with delirium and compared them with dementia and PPD groups. Biomarker concentrations were the primary outcome and were compared using general linear models adjusted for age, sex, and estimated glomerular filtration rate, with bootstrapped confidence intervals. Discriminative performance was assessed as a secondary outcome using receiver operating characteristic (ROC) curve analysis. In Phase 2, we applied this threshold to independent dementia and PPD groups. Results: In Phase 1 (n =176), inpatients with delirium showed higher concentrations of both NfL and GFAP than patients with dementia ({beta} = 0.78; 95% CI [0.37, 1.20] and {beta} = 0.39 [0.07, 0.73] respectively) or PPD groups ({beta} = 1.17; [0.69, 1.66] and {beta} = 0.83 [0.51, 1.24] respectively). ROC curve analysis for differentiating delirium from dementia and PPD groups yielded an optimal NfL threshold of 24.1 pg/mL (AUC = 0.89, 96% sensitivity, 70% specificity). In comparison, GFAP showed lower accuracy (AUC = 0.85), and combining NfL with GFAP (AUC = 0.90) did not significantly improve performance. In Phase 2 (n = 154), the NfL threshold of > 24 pg/ml maintained strong performance (AUC = 0.95, 96% sensitivity, and 77% specificity). Conclusions: Plasma NfL shows promise as a biomarker for delirium in hospitalised patients, distinguishing it with high sensitivity from patients with dementia and PPD. Larger multicentre studies are needed to refine cut-offs, evaluate temporal dynamics, and determine integration into acute hospital delirium pathways.

Background: Intracerebral hemorrhage (ICH) is associated with high early mortality. Although routine coagulation markers are clinically relevant, their incremental value for mortality prediction beyond neurologic and physiologic information is uncertain. We developed and externally validated models using first-24-hour ICU data and evaluated the added predictive value of routine coagulation measurements. Methods: We used MIMIC-IV for model development and eICU-CRD for external validation. The development cohort included 2,669 adults with non-traumatic ICH and 388 ICU deaths. Logistic regression, random forest, and XGBoost were evaluated using a patient-level 70:30 holdout split and 5-fold stratified cross-validation. External validation used a high-specificity primary eICU cohort (396 patients; 39 deaths) and a broader secondary hemorrhagic-stroke/likely-ICH cohort (1,555 patients; 168 deaths). Performance was assessed using AUROC, AUPRC, Brier score, calibration, and paired bootstrap comparisons. GCS-only and age-plus-GCS models served as clinical benchmarks. Coagulation-only, clinical-only, and combined XGBoost models were compared to assess incremental value. Results: In the internal holdout cohort, AUROCs were 0.870 for logistic regression, 0.859 for random forest, and 0.867 for XGBoost. In the primary external cohort, corresponding AUROCs were 0.804, 0.839, and 0.817; AUPRCs were 0.342, 0.419, and 0.406; and Brier scores were 0.080, 0.072, and 0.074. In the broader cohort, AUROCs were 0.826, 0.848, and 0.838. Random forest showed the numerically strongest external performance, but pairwise AUROC differences among multivariable models were not clearly supported. GCS-only achieved AUROCs of 0.780 and 0.813 in the primary and broader cohorts. Coagulation-only XGBoost showed weak discrimination (AUROC 0.598 internally, 0.598 primary external, and 0.566 broader external). Adding coagulation variables to the clinical model produced AUROC differences of 0.014 (95% CI -0.009 to 0.037) and 0.003 (95% CI -0.007 to 0.013) in the two external cohorts. Conclusions: Multivariable models based on first-24-hour ICU data retained useful discrimination across external cohorts and provided modest gains over simple GCS-based benchmarks. Routine coagulation markers alone were weak predictors and did not provide clearly supported incremental discrimination beyond broader clinical information. Cross-database transportability should be interpreted together with phenotype alignment and calibration.

Background. Hematoporphyrin derivative (HpD, Hiporfin) is used for photodynamic therapy (PDT) of solid tumors but had never been evaluated for fluorescence guided resection (FGR) in glioma. We assessed Hiporfin fluorescence accuracy against histopathology in recurrent glioblastoma (rGBM) and the feasibility of adding cavity PDT to FGR. Methods. In this prospective single arm series, adults with rGBM received a single preoperative intravenous Hiporfin dose of 5 mg/kg before FGR under blue light excitation and 630 nm cavity PDT. Intraoperative fluorescence assessment was the index test and histopathology the reference standard. Specimen level diagnostic accuracy was pooled with bivariate random-effects modeling. Results. Of 63 specimens, 35 were true positive, 7 false positive, 8 false negative and 13 true negative. Pooled sensitivity was 0.751 (95% confidence interval 0.600 to 0.858) and specificity 0.606 (0.407 to 0.775); the crude positive predictive value was 0.833 (0.694 to 0.917) and the negative predictive value 0.619 (0.409 to 0.792). All seven false positive specimens were peritumoral tissue and the eight false negatives were non fluorescent tumor. FGR combined with cavity PDT was completed in all six patients without new neurological deficits. Conclusions. Hiporfin fluorescence identified rGBM with moderate accuracy: fluorescent regions were likely tumor, while non fluorescent tissue often still contained tumor. The main source of error was peritumoral tissue of the treated cavity, and not the drug or the optics. Hiporfin is a new fluorescence agent for glioma. One drug provides both fluorescence guidance during resection and immediate cavity PDT.

Background: Alpha-mannosidosis (AM) is a debilitating lysosomal storage disorder (LSD) caused by insufficient activity of alpha-mannosidase, an enzyme responsible for cleavage of alpha-linked mannose residues from N-glycan-type oligosaccharides during their degradation. As a result, specific oligosaccharides accumulate in tissues and are also detectable in body fluids. AM is a progressive multi-organ disease for which hematopoietic stem cell transplantation (HSCT) and enzyme replacement therapy (ERT) are therapeutic options, however, treatment results depend considerably on early diagnosis. Therefore, inclusion of AM into newborn screening panels would significantly contribute to improving therapeutic outcome. Methods: A rapid and fully quantitative method for the specific AM oligosaccharide biomarker Man2GlcNAc in dried blood spots (DBS) using ultra-performance liquid chromatography coupled to tandem-mass spectrometry was developed. DBS from 420 controls, 27 AM patients (including five newborn screening samples), and 35 patients with other LSDs were analyzed. Results: A distinct separation in biomarker concentrations was observed between unaffected individuals and untreated AM patients (0.1 - 0.8 vs 3.2 - 11.3 mol Man2GlcNAc/L, p<0.0001). Furthermore, Man2GlcNAc concentrations significantly differed between untreated patients and those on ERT (n = 17, p<0.0001) or after HSCT (n = 15, p<0.0001), and between both treatment groups (p<0.0001). Moreover, we demonstrated that this method is appropriate for monitoring treatment efficacy by analysis of samples from individual patients before and after initiation of therapy. We retrospectively analyzed residual samples of dried blood spots (DBS) from routine newborn screening that had been stored from patients with later-confirmed AM, which resulted in clear separation from the concentrations observed in a newborn screening control cohort (7.7 - 11.3 vs 0.1 - 0.8 mol Man2GlcNAc/L, p=0.0001). Conclusion: We present a simple and rapid method for quantitative analysis of the specific AM oligosaccharide biomarker Man2GlcNAc, which is remarkably stable compared to enzymatic assays. Thus, the method is useful for selective screening in suspected cases as well as for monitoring the response to different forms of treatment. Furthermore, Man2GlcNAc has the potential to serve as a useful biomarker for newborn screening offering the prospect of earlier treatment initiation, improved therapeutic outcome and higher quality of life for AM patients.

Context: Cytological specimens allow minimally invasive, repeated sampling for diagnosis and molecular testing. However, cytology biobanking protocols have primarily been optimized for morphology, and the effects of preanalytical handling and storage on intracellular proteins remain unclear. Objective: To assess how fixation, staining, storage temperature, and duration affect intracellular protein preservation in cytological specimens and to identify conditions suitable for in situ proximity ligation assay (isPLA). Design: Human cancer cell lines were processed under different preanalytical conditions. Intracellular proteins and protein protein interactions were assessed by isPLA, fluorescence microscopy, and quantitative image analysis. Slides were stored at room temperature or minus 20 Celsius, with or without MayGrunwaldGiemsa (MGG) staining. Results: Storage temperature had the greatest effect on protein signal. Unstained slides stored at minus 20 Celsius retained strong signals for up to 24 months, whereas room temperature storage reduced detectable protein interactions by approximately 95%. MGG staining further reduced signal intensity during prolonged storage. Nevertheless, stained slides remained suitable for isPLA when staining, microscopic evaluation, destaining, and antibody labeling were completed on the same day. Prolonged room temperature storage of stained slides caused loss of antigenicity, increased autofluorescence, and false negative findings. Conclusions: Low temperature storage and standardized preanalytical handling preserve intracellular protein signals in cytological specimens. These conditions support reliable single cell protein profiling of archived samples and their use in biomarker research and translational pathology.