Addressing the Contrast Media Recognition Challenge: A Fully Automated Machine Learning Approach for Predicting Contrast Phases in CT Imaging.

OBJECTIVES: Accurately acquiring and assigning different contrast-enhanced phases in computed tomography (CT) is relevant for clinicians and for artificial intelligence orchestration to select the most appropriate series for analysis. However, this information is commonly extracted from the CT metadata, which is often wrong. This study aimed at developing an automatic pipeline for classifying intravenous (IV) contrast phases and additionally for identifying contrast media in the gastrointestinal tract (GIT). MATERIALS AND METHODS: This retrospective study used 1200 CT scans collected at the investigating institution between January 4, 2016 and September 12, 2022, and 240 CT scans from multiple centers from The Cancer Imaging Archive for external validation. The open-source segmentation algorithm TotalSegmentator was used to identify regions of interest (pulmonary artery, aorta, stomach, portal/splenic vein, liver, portal vein/hepatic veins, inferior vena cava, duodenum, small bowel, colon, left/right kidney, urinary bladder), and machine learning classifiers were trained with 5-fold cross-validation to classify IV contrast phases (noncontrast, pulmonary arterial, arterial, venous, and urographic) and GIT contrast enhancement. The performance of the ensembles was evaluated using the receiver operating characteristic area under the curve (AUC) and 95% confidence intervals (CIs). RESULTS: For the IV phase classification task, the following AUC scores were obtained for the internal test set: 99.59% [95% CI, 99.58-99.63] for the noncontrast phase, 99.50% [95% CI, 99.49-99.52] for the pulmonary-arterial phase, 99.13% [95% CI, 99.10-99.15] for the arterial phase, 99.8% [95% CI, 99.79-99.81] for the venous phase, and 99.7% [95% CI, 99.68-99.7] for the urographic phase. For the external dataset, a mean AUC of 97.33% [95% CI, 97.27-97.35] and 97.38% [95% CI, 97.34-97.41] was achieved for all contrast phases for the first and second annotators, respectively. Contrast media in the GIT could be identified with an AUC of 99.90% [95% CI, 99.89-99.9] in the internal dataset, whereas in the external dataset, an AUC of 99.73% [95% CI, 99.71-99.73] and 99.31% [95% CI, 99.27-99.33] was achieved with the first and second annotator, respectively. CONCLUSIONS: The integration of open-source segmentation networks and classifiers effectively classified contrast phases and identified GIT contrast enhancement using anatomical landmarks.

Optimizing platelet transfusion through a personalized deep learning risk assessment system for demand management.

ABSTRACT: Platelet demand management (PDM) is a resource-consuming task for physicians and transfusion managers of large hospitals. Inpatient numbers and institutional standards play significant roles in PDM. However, reliance on these factors alone commonly results in platelet shortages. Using data from multiple sources, we developed, validated, tested, and implemented a patient-specific approach to support PDM that uses a deep learning-based risk score to forecast platelet transfusions for each hospitalized patient in the next 24 hours. The models were developed using retrospective electronic health record data of 34 809 patients treated between 2017 and 2022. Static and time-dependent features included demographics, diagnoses, procedures, blood counts, past transfusions, hematotoxic medications, and hospitalization duration. Using an expanding window approach, we created a training and live-prediction pipeline with a 30-day input and 24-hour forecast. Hyperparameter tuning determined the best validation area under the precision-recall curve (AUC-PR) score for long short-term memory deep learning models, which were then tested on independent data sets from the same hospital. The model tailored for hematology and oncology patients exhibited the best performance (AUC-PR, 0.84; area under the receiver operating characteristic curve [ROC-AUC], 0.98), followed by a multispecialty model covering all other patients (AUC-PR, 0.73). The model specific to cardiothoracic surgery had the lowest performance (AUC-PR, 0.42), likely because of unexpected intrasurgery bleedings. To our knowledge, this is the first deep learning-based platelet transfusion predictor enabling individualized 24-hour risk assessments at high AUC-PR. Implemented as a decision-support system, deep-learning forecasts might improve patient care by detecting platelet demand earlier and preventing critical transfusion shortages.

FHIR-PYrate: a data science friendly Python package to query FHIR servers.

BACKGROUND: We present FHIR-PYrate, a Python package to handle the full clinical data collection and extraction process. The software is to be plugged into a modern hospital domain, where electronic patient records are used to handle the entire patient's history. Most research institutes follow the same procedures to build study cohorts, but mainly in a non-standardized and repetitive way. As a result, researchers spend time writing boilerplate code, which could be used for more challenging tasks. METHODS: The package can improve and simplify existing processes in the clinical research environment. It collects all needed functionalities into a straightforward interface that can be used to query a FHIR server, download imaging studies and filter clinical documents. The full capacity of the search mechanism of the FHIR REST API is available to the user, leading to a uniform querying process for all resources, thus simplifying the customization of each use case. Additionally, valuable features like parallelization and filtering are included to make it more performant. RESULTS: As an exemplary practical application, the package can be used to analyze the prognostic significance of routine CT imaging and clinical data in breast cancer with tumor metastases in the lungs. In this example, the initial patient cohort is first collected using ICD-10 codes. For these patients, the survival information is also gathered. Some additional clinical data is retrieved, and CT scans of the thorax are downloaded. Finally, the survival analysis can be computed using a deep learning model with the CT scans, the TNM staging and positivity of relevant markers as input. This process may vary depending on the FHIR server and available clinical data, and can be customized to cover even more use cases. CONCLUSIONS: FHIR-PYrate opens up the possibility to quickly and easily retrieve FHIR data, download image data, and search medical documents for keywords within a Python package. With the demonstrated functionality, FHIR-PYrate opens an easy way to assemble research collectives automatically.

Explainable AI in medical imaging: An overview for clinical practitioners - Saliency-based XAI approaches.

Since recent achievements of Artificial Intelligence (AI) have proven significant success and promising results throughout many fields of application during the last decade, AI has also become an essential part of medical research. The improving data availability, coupled with advances in high-performance computing and innovative algorithms, has increased AI's potential in various aspects. Because AI rapidly reshapes research and promotes the development of personalized clinical care, alongside its implementation arises an urgent need for a deep understanding of its inner workings, especially in high-stake domains. However, such systems can be highly complex and opaque, limiting the possibility of an immediate understanding of the system's decisions. Regarding the medical field, a high impact is attributed to these decisions as physicians and patients can only fully trust AI systems when reasonably communicating the origin of their results, simultaneously enabling the identification of errors and biases. Explainable AI (XAI), becoming an increasingly important field of research in recent years, promotes the formulation of explainability methods and provides a rationale allowing users to comprehend the results generated by AI systems. In this paper, we investigate the application of XAI in medical imaging, addressing a broad audience, especially healthcare professionals. The content focuses on definitions and taxonomies, standard methods and approaches, advantages, limitations, and examples representing the current state of research regarding XAI in medical imaging. This paper focuses on saliency-based XAI methods, where the explanation can be provided directly on the input data (image) and which naturally are of special importance in medical imaging.

Explainable AI in medical imaging: An overview for clinical practitioners - Beyond saliency-based XAI approaches.

Driven by recent advances in Artificial Intelligence (AI) and Computer Vision (CV), the implementation of AI systems in the medical domain increased correspondingly. This is especially true for the domain of medical imaging, in which the incorporation of AI aids several imaging-based tasks such as classification, segmentation, and registration. Moreover, AI reshapes medical research and contributes to the development of personalized clinical care. Consequently, alongside its extended implementation arises the need for an extensive understanding of AI systems and their inner workings, potentials, and limitations which the field of eXplainable AI (XAI) aims at. Because medical imaging is mainly associated with visual tasks, most explainability approaches incorporate saliency-based XAI methods. In contrast to that, in this article we would like to investigate the full potential of XAI methods in the field of medical imaging by specifically focusing on XAI techniques not relying on saliency, and providing diversified examples. We dedicate our investigation to a broad audience, but particularly healthcare professionals. Moreover, this work aims at establishing a common ground for cross-disciplinary understanding and exchange across disciplines between Deep Learning (DL) builders and healthcare professionals, which is why we aimed for a non-technical overview. Presented XAI methods are divided by a method's output representation into the following categories: Case-based explanations, textual explanations, and auxiliary explanations.

Gene Augmentation and Readthrough Rescue Channelopathy in an iPSC-RPE Model of Congenital Blindness.

Pathogenic variants of the KCNJ13 gene are known to cause Leber congenital amaurosis (LCA16), an inherited pediatric blindness. KCNJ13 encodes the Kir7.1 subunit that acts as a tetrameric, inwardly rectifying potassium ion channel in the retinal pigment epithelium (RPE) to maintain ionic homeostasis and allow photoreceptors to encode visual information. We sought to determine whether genetic approaches might be effective in treating blindness arising from pathogenic variants in KCNJ13. We derived human induced pluripotent stem cell (hiPSC)-RPE cells from an individual carrying a homozygous c.158G>A (p.Trp53∗) pathogenic variant of KCNJ13. We performed biochemical and electrophysiology assays to confirm Kir7.1 function. We tested both small-molecule readthrough drug and gene-therapy approaches for this "disease-in-a-dish" approach. We found that the LCA16 hiPSC-RPE cells had normal morphology but did not express a functional Kir7.1 channel and were unable to demonstrate normal physiology. After readthrough drug treatment, the LCA16 hiPSC cells were hyperpolarized by 30 mV, and the Kir7.1 current was restored. Similarly, we rescued Kir7.1 channel function after lentiviral gene delivery to the hiPSC-RPE cells. In both approaches, Kir7.1 was expressed normally, and there was restoration of membrane potential and the Kir7.1 current. Loss-of-function variants of Kir7.1 are one cause of LCA. Using either readthrough therapy or gene augmentation, we rescued Kir7.1 channel function in iPSC-RPE cells derived from an affected individual. This supports the development of precision-medicine approaches for the treatment of clinical LCA16.

Multiple Dictyostelid Species Destroy Biofilms of Klebsiella oxytoca and Other Gram Negative Species.

Dictyostelids are free-living phagocytes that feed on bacteria in diverse habitats. When bacterial prey is in short supply or depleted, they undergo multicellular development culminating in the formation of dormant spores. In this work, we tested isolates representing four dictyostelid species from two genera (Dictyostelium and Polysphondylium) for the potential to feed on biofilms preformed on glass and polycarbonate surfaces. The abilities of dictyostelids were monitored for three hallmarks of activity: 1) spore germination on biofilms, 2) predation on biofilm enmeshed bacteria by phagocytic cells and 3) characteristic stages of multicellular development (streaming and fructification). We found that all dictyostelid isolates tested could feed on biofilm enmeshed bacteria produced by human and plant pathogens: Klebsiella oxytoca, Pseudomonas aeruginosa, Pseudomonas syringae, Erwinia amylovora 1189 (biofilm former) and E. amylovora 1189 Δams (biofilm deficient mutant). However, when dictyostelids were fed planktonic E. amylovora Δams the bacterial cells exhibited an increased susceptibility to predation by one of the two dictyostelid strains they were tested against. Taken together, the qualitative and quantitative data presented here suggest that dictyostelids have preferences in bacterial prey which affects their efficiency of feeding on bacterial biofilms.

Regulation of WNT Signaling by VSX2 During Optic Vesicle Patterning in Human Induced Pluripotent Stem Cells.

Few gene targets of Visual System Homeobox 2 (VSX2) have been identified despite its broad and critical role in the maintenance of neural retina (NR) fate during early retinogenesis. We performed VSX2 ChIP-seq and ChIP-PCR assays on early stage optic vesicle-like structures (OVs) derived from human iPS cells (hiPSCs), which highlighted WNT pathway genes as direct regulatory targets of VSX2. Examination of early NR patterning in hiPSC-OVs from a patient with a functional null mutation in VSX2 revealed mis-expression and upregulation of WNT pathway components and retinal pigmented epithelium (RPE) markers in comparison to control hiPSC-OVs. Furthermore, pharmacological inhibition of WNT signaling rescued the early mutant phenotype, whereas augmentation of WNT signaling in control hiPSC-OVs phenocopied the mutant. These findings reveal an important role for VSX2 as a regulator of WNT signaling and suggest that VSX2 may act to maintain NR identity at the expense of RPE in part by direct repression of WNT pathway constituents. Stem Cells 2016;34:2625-2634.

VSX2 and ASCL1 Are Indicators of Neurogenic Competence in Human Retinal Progenitor Cultures.

Three dimensional (3D) culture techniques are frequently used for CNS tissue modeling and organoid production, including generation of retina-like tissues. A proposed advantage of these 3D systems is their potential to more closely approximate in vivo cellular microenvironments, which could translate into improved manufacture and/or maintenance of neuronal populations. Visual System Homeobox 2 (VSX2) labels all multipotent retinal progenitor cells (RPCs) and is known to play important roles in retinal development. In contrast, the proneural transcription factor Acheate scute-like 1 (ASCL1) is expressed transiently in a subset of RPCs, but is required for the production of most retinal neurons. Therefore, we asked whether the presence of VSX2 and ASCL1 could gauge neurogenic potential in 3D retinal cultures derived from human prenatal tissue or ES cells (hESCs). Short term prenatal 3D retinal cultures displayed multiple characteristics of human RPCs (hRPCs) found in situ, including robust expression of VSX2. Upon initiation of hRPC differentiation, there was a small increase in co-labeling of VSX2+ cells with ASCL1, along with a modest increase in the number of PKCα+ neurons. However, 3D prenatal retinal cultures lost expression of VSX2 and ASCL1 over time while concurrently becoming refractory to neuronal differentiation. Conversely, 3D optic vesicles derived from hESCs (hESC-OVs) maintained a robust VSX2+ hRPC population that could spontaneously co-express ASCL1 and generate photoreceptors and other retinal neurons for an extended period of time. These results show that VSX2 and ASCL1 can serve as markers for neurogenic potential in cultured hRPCs. Furthermore, unlike hESC-OVs, maintenance of 3D structure does not independently convey an advantage in the culture of prenatal hRPCs, further illustrating differences in the survival and differentiation requirements of hRPCs extracted from native tissue vs. those generated entirely in vitro.