Biology subjects

Wong, K. A.

Publications and source records attributed to Wong, K. A..

4 recordsLinked to original sources

bioRxiv2026

CALM-VLM: CALIBRATION AND SELECTIVE PREDICTION IN VISION-LANGUAGE MODELS FOR RELIABLE BRAIN MRI CLASSIFICATION

AO_SCPLOWBSTRACTC_SCPLOWRecent advances in vision-language models (VLMs) have demonstrated strong multimodal capabilities for medical image analysis. However, their confidence in diagnostic predictions is often unclear, limiting adoption in clinical settings. We introduce CALM-VLM (CALibration Mechanism for Vision-Language Models), which integrates confidence calibration and selective prediction into a generative 3D VLM. To create CALM-VLM, we fine-tuned the Med3DVLM architecture for Alzheimers disease (AD) and stroke classification as initial test cases. To improve reliability, we incorporated temperature scaling on the VLMs generative outputs. The calibrated model then selectively abstained from predictions when uncertain; this also improved its diagnostic accuracy. Experiments across multi-site MRI datasets, from 10 countries worldwide, show that CALM-VLM improved confidence relative to uncalibrated VLMs. Coverage-adjusted test receiver-operator characteristic curve-area under the curve (ROC-AUC) increased by 5% to 13% for both diagnostic tasks across independent test sets. Our calibrated VLM achieved a test ROC-AUC of 0.951 for AD classification and 0.905 for stroke classification. These findings highlight the importance of calibrated, uncertainty-aware VLMs for trustworthy neuroimaging AI.

neuroscience↗

bioRxiv2026

Ultra-large targeted DNA integrations in primary human cells

Genetic engineering experiments and therapies are constrained by the size of DNA integrations into human cells genomes. Existing AAV, lentiviral, and non-viral methods rapidly decrease in integration efficiency beyond [~]5kb of sequence. Through systematic evaluation of non-viral DNA template formats, we identified circular ssDNA and dsDNA as capable of mediating >5kb integrations. Large circular DNA delivery efficiency and its impacts on cell viability and payload expression could be significantly improved with small DNA "helper" plasmids, mRNA-encoded nucleases, and sequence design optimizations. Collectively, these modifications enabled ultra-large--up to 10 kb DNA--integrations at >20% efficiency in primary human T cells at the TRAC locus and at >60% efficiency in human iPSCs at the AAVS1 locus. Finally, we demonstrate that GMP clinical-manufactured T cells with ultra-large integrations are functional in vitro and in vivo. Overall, we identified optimal template architectures, delivery modes, and sequence design rules for ultra-large DNA integrations in both research and clinical settings to accelerate basic genetic research and next-generation cellular therapies.

genetics↗

bioRxiv2023

Data-driven biomarkers outperform theory-based biomarkers in predicting stroke motor outcomes

Chronic motor impairments are a leading cause of disability after stroke. Previous studies have predicted motor outcomes based on the degree of damage to predefined structures in the motor system, such as the corticospinal tract. However, such theory-based approaches may not take full advantage of the information contained in clinical imaging data. The present study uses data-driven approaches to predict chronic motor outcomes after stroke and compares the accuracy of these predictions to previously-identified theory-based biomarkers. Using a cross-validation framework, regression models were trained using lesion masks and motor outcomes data from 789 stroke patients (293 female/496 male) from the ENIGMA Stroke Recovery Working Group (age 64.9{+/-}18.0 years; time since stroke 12.2{+/-}0.2 months; normalised motor score 0.7{+/-}0.5 (range [0,1]). The out-of-sample prediction accuracy of two theory-based biomarkers was assessed: lesion load of the corticospinal tract, and lesion load of multiple descending motor tracts. These theory-based prediction accuracies were compared to the prediction accuracy from three data-driven biomarkers: lesion load of lesion-behaviour maps, lesion load of structural networks associated with lesion-behaviour maps, and measures of regional structural disconnection. In general, data-driven biomarkers had better prediction accuracy - as measured by higher explained variance in chronic motor outcomes - than theory-based biomarkers. Data-driven models of regional structural disconnection performed the best of all models tested (R2 = 0.210, p < 0.001), performing significantly better than predictions using the theory-based biomarkers of lesion load of the corticospinal tract (R2 = 0.132, p< 0.001) and of multiple descending motor tracts (R2 = 0.180, p < 0.001). They also performed slightly, but significantly, better than other data-driven biomarkers including lesion load of lesion-behaviour maps (R2 =0.200, p < 0.001) and lesion load of structural networks associated with lesion-behaviour maps (R2 =0.167, p < 0.001). Ensemble models - combining basic demographic variables like age, sex, and time since stroke - improved prediction accuracy for theory-based and data-driven biomarkers. Finally, combining both theory-based and data-driven biomarkers with demographic variables improved predictions, and the best ensemble model achieved R2 = 0.241, p < 0.001. Overall, these results demonstrate that models that predict chronic motor outcomes using data-driven features, particularly when lesion data is represented in terms of structural disconnection, perform better than models that predict chronic motor outcomes using theory-based features from the motor system. However, combining both theory-based and data-driven models provides the best predictions.

neuroscience↗

bioRxiv2023

Complement C3 interacts with cytochrome c to influence myocardial apoptosis during heart ischemia/reperfusion

Myocardial ischemia/reperfusion (I/R) elicits an acute inflammatory response involving complement factors. Previous animal studies showed that circulation complement C3 was deposited in the ischemic myocardium flooded with oxygenated blood upon reperfusion. Recently, we reported that myocardial necrosis was decreased in C3-/- mice after heart I/R. The current study used in the same heart model to test the effect of C3 on myocardial apoptosis. Our results showed that myocardial apoptosis was increased in C3-/- mice after heart I/R. Further, comparative proteomics analyses found that cytochrome c was present in the myocardial C3-complex following I/R. These results indicate that C3 can interact with cytochrome c in the cytosol of cardiomyocytes during myocardial I/R, which may sequester cytochrome c and thus reduce the number of cells undergoing apoptosis. In summary, our findings raise the possibility of a new mechanism affecting cell death relevant to pathologic conditions such as ischemia: a circulating innate immune factor, i.e. complement, can interact with intracellular factor(s), and influence the types of cell death that occur.

molecular biology↗