Stefano Casarin

Background: Candida species distribution has changed and antifungal resistance has increased since COVID-19. We determined whether these trends contributed to changes in candidemia mortality. Methods: In a retrospective cohort study of patients with candidemia in Houston, TX, we compared clinical characteristics, species, and resistance between 3 periods (P1, 2017–2019; P2, 2020–2022; and P3, 2023–2025) and assessed contribution to mortality. We compared 30-day mortality in candidemia versus bacteremia. Results: Among 1275 patients with candidemia (1386 episodes), 30-day mortality increased from P1 to P2 (19.7% vs 31.9%, P < .001) and remained above baseline in P3 (29.8%, P < .001 vs P1). In contrast, 30-day mortality increased from P1 to P2 in patients with Enterobacterales (7.4% vs 9.5%, P = .004) and Staphylococcus aureus bacteremia (9.5% vs 14.3%, P < .001) but returned to baseline in P3 (both P > .05 vs P1). Mean age (62.7 ± 16.0 vs 66.1 ± 14.8, P = .002) and median Elixhauser Comorbidity Index (8 [IQR 5–12] vs 11 [IQR 6–15], P < .001) increased from P1 to P3 among candidemia patients. Proportion of ICU-onset candidemia increased from P1 to P2 (31.3% vs 40.6%, P = .02). Proportion of C. auris increased from P1 to P3 (0.5% vs 20.5%, P < .001). In multivariable analyses, age (OR 1.03, 95% CI 1.02–1.04) and ICU-onset candidemia (OR 2.08, 95%CI 1.46–2.98) were associated with increased mortality. In univariable analyses, C. auris and antifungal resistance were not associated with increased mortality. Conclusions: Increased post-candidemia mortality was associated in part with demographic changes and not with species or resistance changes. Factors underlying the persistent increase in post-candidemia mortality are unclear.

Background and objectives: Bone is a multicellular organ that is the site of complex pathophysiological events (such as cancer). 3D confocal and multiphoton microscopy, followed by manual analysis and quantification, allow the identification of molecular, cellular, and tissue mechanisms in their original context, enabling the investigation of spatial biology at subcellular level. The automation of these analyses is becoming increasingly important due to their time-consuming nature, lack of standardization, and inter-subject variability. Methods: Deep learning applied to image analysis could overcome current limitations and optimize preclinical research. This includes automatic semantic segmentation of bone cells (osteoblasts, osteoclasts, and blood vessels) and mineral component, followed by parameter extraction and quantification. Accordingly, fluorescence microscopy images were generated, pre-processed, and fed (total 21,395 images: 17,104 for training, 4291 for validation) into a neural network-based architecture named Bo-Net. Results: The fully automatic Bo-Net achieved strong performance across the tested metrics and segmentation accuracy similar to experienced biologists (R = 0.81–0.99 for parameters tested) and improved analysis time from days to seconds in a variety of experimental conditions, confirming its relevance in different biologically relevant contexts. Conclusions: As a result, a fully automated, yet reliable, tool to optimize the analysis of multiparametric fluorescence microscopy images available to the bone research community was generated.

Modèles mathématiques en chirurgie informatisée: La chirurgie informatisée est une science nouvelle dont le but est de croiser la chirurgie avec les sciences de l’informatique afin d’aboutir à des améliorations significatives dans les deux domaines. Avec l’évolution des nouvelles techniques chirurgicales, une collaboration étroite entre chirurgiens et chercheurs est devenue à la fois inévitable et essentielle à l’optimisation des soins chirurgicaux. L’utilisation de modèles mathématiques est la pierre angulaire de ce nouveau domaine. Cette thèse démontre comment une approche systématique d’un problème clinique nous a amenés à répondre à des questions ouvertes dans le domaine chirurgical en utilisant des modèles mathématiques à grande échelle. De manière générale, notre approche inclut (i) une vision générale du problème, (ii) le ciblage du/des système(s) physiologique(s) à étudier pour y répondre, et (iii) un effort de modélisation mathématique, qui a toujours été poussé par la recherche d’un compromis entre complexité du système étudié et réalité physiologique. Nous avons consacré la première partie de cette thèse à l’optimisation des conditions limites à appliquer à un bio-réacteur utilisé pour démultiplier le tissu pulmonaire provenant d’un donneur. Un modèle géométrique de l’arbre trachéo-bronchique couplé à un modèle de dépôt de soluté nous a permis de déterminer l’ensemble des pressions à appliquer aux pompes servant le bio-réacteur afin d’obtenir une distribution optimale des nutriments à travers les cultures de tissus. Nous avons consacré la seconde partie de cette thèse au problème de resténose des greffes de veines utilisées pour contourner une occlusion artérielle. Nous avons reproduit l’apparition de resténose grâce à plusieurs modèles mathématiques qui permettent d’étudier les preuves cliniques et de tester des hypothèses cliniques avec un niveau croissant de complexité et de précision. Pour finir, nous avons développé un cadre de travail robuste pour tester les effets des thérapies géniques afin de limiter la resténose. Une découverte intéressante a été de constater qu’en contrôlant un groupe de gènes spécifique, la perméabilité à la lumière double après un mois de suivi. Grace aux résultats obtenus, nous avons démontré que la modélisation mathématique peut servir de puissant outil pour l’innovation chirurgicale.

Chronic rejection remains a significant challenge in solid organ transplantation, contributing to graft dysfunction and eventual failure despite advances in immunosuppressive therapies. Computational modeling has emerged as a powerful tool for understanding chronic rejection mechanisms, enhancing diagnostic precision, and identifying novel therapeutic targets. This chapter explores various computational approaches, including artificial intelligence, machine learning, ordinary differential equations, partial differential equations, agent-based models, and gene network analysis applied to solid organ transplantation: kidney, liver, heart, and lung. While computational models offer numerous advantages, including cost-effectiveness and the ability to integrate multi-omics data, challenges remain in terms of data quality, standardization, and clinical validation. Bridging these gaps will require comprehensive longitudinal studies and the development of hybrid models that combine diverse computational techniques. Emerging technologies such as single-cell transcriptomics and spatial genomics hold promise for enhancing predictive accuracy and understanding cellular heterogeneity. As computational methods evolve, their integration with experimental research will be essential for developing precision medicine strategies to improve long-term graft survival.

Objectives: Candida species are one of the most common causes of ICU-onset bloodstream infection (BSI). Yet, there is no robust guidance on when to initiate empiric antifungals for ICU patients suspected of BSI. We compared patients with ICU-onset candidemia vs. ICU-onset bacteremia to determine which patients may benefit from empiric antifungals. Design: We compared characteristics between patients with bacteremia vs. candidemia at time of culture. We determined risk factors for death and constructed a multivariable regression to determine if candidemia is an independent risk factor for death. Setting/Patients: We included ICU patients from a hospital system in Houston, Texas ( n = 8 hospitals) with a BSI from 2016 to 2023 and validated our findings using ICU patients with a BSI in the publicly available Medical Information Mart for Intensive Care (MIMIC)-IV cohort. Interventions: None. Measurements and Main Results: Of 1509 total patients in our primary cohort, 290 (19.2%) had candidemia and 1219 (80.8%) had bacteremia. Patients with candidemia were more likely to be on invasive mechanical ventilation (72.8% vs. 52.8%; p < 0.001), vasopressors (39.3% vs. 24.0%; p < 0.001), and continuous renal replacement therapy (15.2% vs. 9.4%; p = 0.006) at the time of culture. They contracted infection later in ICU stay than bacteremia patients (7.2 vs. 5.0 d; p < 0.001) and were more likely to die within 30 days of culture (unadjusted odds ratio [OR], 1.62; 95% CI, 1.25–2.09). After adjusting for ICU interventions (invasive mechanical ventilation, vasopressors, and continuous renal replacement therapy) and baseline parameters, candidemia was not independently associated with mortality compared with bacteremia in our primary cohort (OR, 1.21; 95% CI, 0.92–1.60) but was in the MIMIC-IV cohort (OR, 1.48; 95% CI, 1.003–2.17). Conclusions: Patients with significant ICU resource utilization are at increased risk for candidemia. Our data suggest that when initiating empiric antibiotics in patients requiring high-resource ICU care, empiric antifungal therapy should be considered.

Implementation of bone stromal cells (acquisition by MPM and implementation in the A(BM)2)

Impact of tumor positioning on bone resorption, representative pictures

Bone metastasis (BM) is a leading cause of morbidity and mortality in patients with prostate and renal cancer. The complex and dynamic biological processes driving its progression present significant challenges for both understanding and treating this disease. Although current in vivo research provides valuable insights, it is often limited by the inability to fully capture the intricate and multifactorial nature of BM. Thus, complementing existing in vivo models with multiscale computational approaches is crucial for dissecting the complex interactions between tumor cells and the bone microenvironment to advance our understanding of the metastatic process and therapy response. Accordingly, we developed a series of in vivo-inspired, spatially explicit, multicellular agent-based models of BM that effectively recapitulate key aspects of tumor progression, including angiogenesis and bone resorption. The digital twins were rigorously calibrated using in vivo data from prostate and kidney tumors. The models have utility for evaluating therapy response, as verified by the simulation of both the antiangiogenic effects of cabozantinib and the antiresorptive effects of zoledronic acid. These results highlight the predictive character of the agent-based models of BM in anticipating therapeutic outcomes and increasing our understanding of the complex dynamics of BM.Significance:The model offers a transformative tool for evaluating treatment strategies, including combination therapies, with the potential to accelerate the development of highly effective, targeted interventions for bone metastatic cancers.This article is part of a special series: Driving Cancer Discoveries with Computational Research, Data Science, and Machine Learning/AI.