The potential of federated learning for self-configuring medical object detection in heterogeneous data distributions.

Medical Object Detection (MOD) is a clinically relevant image processing method that locates structures of interest in radiological image data at object-level using bounding boxes. High-performing MOD models necessitate large datasets accurately reflecting the feature distribution of the corresponding problem domain. However, strict privacy regulations protecting patient data often hinder data consolidation, negatively affecting the performance and generalization of MOD models. Federated Learning (FL) offers a solution by enabling model training while the data remain at its original source institution. While existing FL solutions for medical image classification and segmentation demonstrate promising performance, FL for MOD remains largely unexplored. Motivated by this lack of technical solutions, we present an open-source, self-configuring and task-agnostic federated MOD framework. It integrates the FL framework Flower with nnDetection, a state-of-the-art MOD framework and provides several FL aggregation strategies. Furthermore, we evaluate model performance by creating simulated Independent Identically Distributed (IID) and non-IID scenarios, utilizing the publicly available datasets. Additionally, a detailed analysis of the distributions and characteristics of these datasets offers insights into how they can impact performance. Our framework's implementation demonstrates the feasibility of federated self-configuring MOD in non-IID scenarios and facilitates the development of MOD models trained on large distributed databases.

AI-Supported Comprehensive Detection and Quantification of Biomarkers of Subclinical Widespread Diseases at Chest CT for Preventive Medicine.

Automated image analysis plays an increasing role in radiology in detecting and quantifying image features outside of the perception of human eyes. Common AI-based approaches address a single medical problem, although patients often present with multiple interacting, frequently subclinical medical conditions. A holistic imaging diagnostics tool based on artificial intelligence (AI) has the potential of providing an overview of multi-system comorbidities within a single workflow. An interdisciplinary, multicentric team of medical experts and computer scientists designed a pipeline, comprising AI-based tools for the automated detection, quantification and characterization of the most common pulmonary, metabolic, cardiovascular and musculoskeletal comorbidities in chest computed tomography (CT). To provide a comprehensive evaluation of each patient, a multidimensional workflow was established with algorithms operating synchronously on a decentralized Joined Imaging Platform (JIP). The results of each patient are transferred to a dedicated database and summarized as a structured report with reference to available reference values and annotated sample images of detected pathologies. Hence, this tool allows for the comprehensive, large-scale analysis of imaging-biomarkers of comorbidities in chest CT, first in science and then in clinical routine. Moreover, this tool accommodates the quantitative analysis and classification of each pathology, providing integral diagnostic and prognostic value, and subsequently leading to improved preventive patient care and further possibilities for future studies.

Interactive Machine Learning-Based Multi-Label Segmentation of Solid Tumors and Organs.

We seek the development and evaluation of a fast, accurate, and consistent method for general-purpose segmentation, based on interactive machine learning (IML). To validate our method, we identified retrospective cohorts of 20 brain, 50 breast, and 50 lung cancer patients, as well as 20 spleen scans, with corresponding ground truth annotations. Utilizing very brief user training annotations and the adaptive geodesic distance transform, an ensemble of SVMs is trained, providing a patient-specific model applied to the whole image. Two experts segmented each cohort twice with our method and twice manually. The IML method was faster than manual annotation by 53.1% on average. We found significant (p < 0.001) overlap difference for spleen (DiceIML/DiceManual = 0.91/0.87), breast tumors (DiceIML/DiceManual = 0.84/0.82), and lung nodules (DiceIML/DiceManual = 0.78/0.83). For intra-rater consistency, a significant (p = 0.003) difference was found for spleen (DiceIML/DiceManual = 0.91/0.89). For inter-rater consistency, significant (p < 0.045) differences were found for spleen (DiceIML/DiceManual = 0.91/0.87), breast (DiceIML/DiceManual = 0.86/0.81), lung (DiceIML/DiceManual = 0.85/0.89), the non-enhancing (DiceIML/DiceManual = 0.79/0.67) and the enhancing (DiceIML/DiceManual = 0.79/0.84) brain tumor sub-regions, which, in aggregation, favored our method. Quantitative evaluation for speed, spatial overlap, and consistency, reveals the benefits of our proposed method when compared with manual annotation, for several clinically relevant problems. We publicly release our implementation through CaPTk (Cancer Imaging Phenomics Toolkit) and as an MITK plugin.

Involvement of hyaluronidases in colorectal cancer.

BACKGROUND: Hyaluronidases belong to a class of enzymes that degrade, predominantly, hyaluronan. These enzymes are known to be involved in physiological and pathological processes, such as tumor growth, infiltration and angiogenesis, but their exact role in tumor promotion or suppression is not clear yet. Advanced colorectal cancer is associated with elevated amounts of hyaluronan of varying size. The aim of the present study was therefore to illuminate the importance of hyaluronidases in colon carcinoma progression. METHODS: The patients' samples (macroscopically normal and cancerous) were subjected to sequential extraction with PBS, 4 M GdnHCl and 4 M GdnHCl --1% Triton X-100. The presence of the various hyaluronidases in the extracts was examined by zymography and western blotting. Their expression was also examined by RT-PCR. RESULTS: Among hyaluronidases examined, Hyal-1, -2, -3 and PH-20 were detected. Their activity was higher in cancerous samples. Hyal-1 and Hyal-2 were overexpressed in cancerous samples, especially in advanced stages of cancer. Both isoforms were mainly extracted with PBS. Hyal-3 was observed only in the third extract of advanced stages of cancer. PH-20 was abundant in all three extracts of all stages of cancer. The expression of only Hyal-1 and PH-20 was verified by RT-PCR. CONCLUSION: A high association of hyaluronidases in colorectal cancer was observed. Each hyaluronidase presented different tissue distribution, which indicated the implication of certain isoforms in certain cancer stages. The results provided new evidence on the mechanisms involved in the progression of colorectal cancer.

The Cancer Imaging Phenomics Toolkit (CaPTk): Technical Overview.

The purpose of this manuscript is to provide an overview of the technical specifications and architecture of the Cancer imaging Phenomics Toolkit (CaPTk www.cbica.upenn.edu/captk), a cross-platform, open-source, easy-to-use, and extensible software platform for analyzing 2D and 3D images, currently focusing on radiographic scans of brain, breast, and lung cancer. The primary aim of this platform is to enable swift and efficient translation of cutting-edge academic research into clinically useful tools relating to clinical quantification, analysis, predictive modeling, decision-making, and reporting workflow. CaPTk builds upon established open-source software toolkits, such as the Insight Toolkit (ITK) and OpenCV, to bring together advanced computational functionality. This functionality describes specialized, as well as general-purpose, image analysis algorithms developed during active multi-disciplinary collaborative research studies to address real clinical requirements. The target audience of CaPTk consists of both computational scientists and clinical experts. For the former it provides i) an efficient image viewer offering the ability of integrating new algorithms, and ii) a library of readily-available clinically-relevant algorithms, allowing batch-processing of multiple subjects. For the latter it facilitates the use of complex algorithms for clinically-relevant studies through a user-friendly interface, eliminating the prerequisite of a substantial computational background. CaPTk's long-term goal is to provide widely-used technology to make use of advanced quantitative imaging analytics in cancer prediction, diagnosis and prognosis, leading toward a better understanding of the biological mechanisms of cancer development.

Multi-institutional noninvasive in vivo characterization of IDH, 1p/19q, and EGFRvIII in glioma using neuro-Cancer Imaging Phenomics Toolkit (neuro-CaPTk).

BACKGROUND: Gliomas represent a biologically heterogeneous group of primary brain tumors with uncontrolled cellular proliferation and diffuse infiltration that renders them almost incurable, thereby leading to a grim prognosis. Recent comprehensive genomic profiling has greatly elucidated the molecular hallmarks of gliomas, including the mutations in isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2), loss of chromosomes 1p and 19q (1p/19q), and epidermal growth factor receptor variant III (EGFRvIII). Detection of these molecular alterations is based on ex vivo analysis of surgically resected tissue specimen that sometimes is not adequate for testing and/or does not capture the spatial tumor heterogeneity of the neoplasm. METHODS: We developed a method for noninvasive detection of radiogenomic markers of IDH both in lower-grade gliomas (WHO grade II and III tumors) and glioblastoma (WHO grade IV), 1p/19q in IDH-mutant lower-grade gliomas, and EGFRvIII in glioblastoma. Preoperative MRIs of 473 glioma patients from 3 of the studies participating in the ReSPOND consortium (collection I: Hospital of the University of Pennsylvania [HUP: n = 248], collection II: The Cancer Imaging Archive [TCIA; n = 192], and collection III: Ohio Brain Tumor Study [OBTS, n = 33]) were collected. Neuro-Cancer Imaging Phenomics Toolkit (neuro-CaPTk), a modular platform available for cancer imaging analytics and machine learning, was leveraged to extract histogram, shape, anatomical, and texture features from delineated tumor subregions and to integrate these features using support vector machine to generate models predictive of IDH, 1p/19q, and EGFRvIII. The models were validated using 3 configurations: (1) 70-30% training-testing splits or 10-fold cross-validation within individual collections, (2) 70-30% training-testing splits within merged collections, and (3) training on one collection and testing on another. RESULTS: These models achieved a classification accuracy of 86.74% (HUP), 85.45% (TCIA), and 75.15% (TCIA) in identifying EGFRvIII, IDH, and 1p/19q, respectively, in configuration I. The model, when applied on combined data in configuration II, yielded a classification success rate of 82.50% in predicting IDH mutation (HUP + TCIA + OBTS). The model when trained on TCIA dataset yielded classification accuracy of 84.88% in predicting IDH in HUP dataset. CONCLUSIONS: Using machine learning algorithms, high accuracy was achieved in the prediction of IDH, 1p/19q, and EGFRvIII mutation. Neuro-CaPTk encompasses all the pipelines required to replicate these analyses in multi-institutional settings and could also be used for other radio(geno)mic analyses.

Alterations of glycosaminoglycan disaccharide content and composition in colorectal cancer: structural and expressional studies.

The glycosaminoglycans are implicated in many processes important in the growth and progression of malignant tumors. In the present study glycosaminoglycans were purified from healthy, macroscopically normal and cancerous specimens of different anatomic sites and different stages of cancer and analyzed by FACE after chondroitinases and sulfatases digestion. The cancerous samples contained increased levels of 6-sulfated unsaturated disaccharides compared to macroscopically normal and healthy samples, the increase being stage-related. The differences in sulfation were found to be related to the anatomic site and the stage of cancer. RT-PCR analysis of 4-sulfotransferase mRNA revealed its presence in decreasing amounts as the stage of the cancer increased. Furthermore, the percent content of hyaluronan disaccharides was elevated in macroscopically normal samples compared to the cancerous, and in addition, it was much more elevated than that of healthy samples. Haluronan levels increase with stage in cancerous tissues. Therefore, it could be concluded that the glycosaminoglycans in colorectal cancer are biosynthetically directed to contribute in different ways depending on the cancer stage and anatomical site.

ADAMTS expression in colorectal cancer.

ADAMTSs are a family of secreted proteinases that share the metalloproteinase domain with matrix metalloproteinases (MMPs). By acting on a large panel of extracellular substrates, they control several cell functions such as fusion, adhesion, proliferation and migration. Through their thrombospondin motifs they also possess anti-angiogenic properties. We investigated whether ADAMTSs participate in colorectal cancer progression and invasion. Their expression was investigated at both mRNA and protein levels. Using RT-PCR, the expression of ADAMTS-1, -4, -5 and ADAMTS-20 was estimated in colorectal tumors of different cancer stage and anatomic site and 3 cell lines of different aggressiveness. An overexpression of ADAMTS-4 and -5 was observed, especially in tissue samples, whereas ADAMTS-1 and -20 were found to be down-regulated. Western blot analysis further supported the RT-PCR findings, revealing in addition the degradation of ADAMTS-1 and -20 in cancer. In situ expression and localization of ADAMTS-1, -4, -5 and -20 was also investigated by immunohistochemical analysis. Our data suggest a positive correlation between ADAMTS-4 and -5 expression and cancer progression, in contrast with the anti-angiogenic members of the family, ADAMTS-1 and -20, which were found to be down-regulated. Our findings support the notion that overexpression of ADAMTS-4 and ADAMTS-5 in colorectal cancer might be a possible invasive mechanism of cancer cells in order to degrade proteoglycans of ECM.

Chondroitin synthases I, II, III and chondroitin sulfate glucuronyltransferase expression in colorectal cancer.

Glycosaminoglycans undergo significant structural alterations in cancer, namely in terms of their sulfation pattern and hydrodynamic size. Numerous studies have focused on this issue, and have demonstrated that glycosaminoglycans play a crucial role in cancer growth and invasion. However, the majority of the enzymes involved in glycosaminoglycan alterations have yet to be examined in detail. The present study focused on the expression of chondroitin-synthesizing enzymes in colorectal cancer. Specimens from healthy controls and cancer patients were subjected to RT-PCR analysis after RNA isolation, and to Western blotting after sequential extraction. The results indicated that chondroitin polymerizing factor and glucuronyltransferase gradually increased with cancer stage, and were expressed at much higher levels in adenomas compared to adjacent normal tissue. The opposite profile was obtained for chondroitin synthase I. Chondroitin synthase III was present at low levels in all the samples examined; however, its expression was higher in the samples from the cancer patients than in those from the healthy controls. It can therefore be concluded that, among the various factors regulating the structure of glycosaminoglycans in cancer, the differential expression of chondroitin-synthesizing enzymes is of the most significance.