Automated Analysis of Split Kidney Function from CT Scans Using Deep Learning and Delta Radiomics.

Background: Differential kidney function assessment is an important part of preoperative evaluation of various urological interventions. It is obtained through dedicated nuclear medical imaging and is not yet implemented through conventional Imaging. Objective: We assess if differential kidney function can be obtained through evaluation of contrast-enhanced computed tomography(CT) using a combination of deep learning and (2D and 3D) radiomic features. Methods: All patients who underwent kidney nuclear scanning at Mayo Clinic sites between 2018-2022 were collected. CT scans of the kidneys were obtained within a 3-month interval before or after the nuclear scans were extracted. Patients who underwent a urological or radiological intervention within this time frame were excluded. A segmentation model was used to segment both kidneys. 2D and 3D radiomics features were extracted and compared between the two kidneys to compute delta radiomics and assess its ability to predict differential kidney function. Performance was reported using receiver operating characteristics, sensitivity, and specificity. Results: Studies from Arizona & Rochester formed our internal dataset (n = 1,159). Studies from Florida were separately processed as an external test set to validate generalizability. We obtained 323 studies from our internal sites and 39 studies from external sites. The best results were obtained by a random forest model trained on 3D delta radiomics features. This model achieved an area under curve (AUC) of 0.85 and 0.81 on internal and external test sets, while specificity and sensitivity were 0.84,0.68 on the internal set, 0.70, and 0.65 on the external set. Conclusion: This proposed automated pipeline can derive important differential kidney function information from contrast-enhanced CT and reduce the need for dedicated nuclear scans for early-stage differential kidney functional assessment. Clinical Impact: We establish a machine learning methodology for assessing differential kidney function from routine CT without the need for expensive and radioactive nuclear medicine scans.

Impact of tertiary lymphoid structures on prognosis and therapeutic response in pancreatic ductal adenocarcinoma.

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) is known to have a heterogeneous desmoplastic tumour microenvironment (TME) with a large number of immunosuppressive cells. Recently, high B-cell infiltration in PDAC has received growing interest as a potential therapeutic target. METHODS: Our literature review summarises the characteristics of tumour-associated tertiary lymphoid structures (TLSs) and highlight the key studies exploring the clinical outcomes of TLSs in PDAC patients and the direct effect on the TME. RESULTS: The location, density and maturity stages of TLSs within tumours play a key role in determining the prognosis and is a new emerging target in cancer immunotherapy. DISCUSSION: TLS development is imperative to improve the prognosis of PDAC patients. In the future, studying the genetics and immune characteristics of tumour infiltrating B cells and TLSs may lead towards enhancing adaptive immunity in PDAC and designing personalised therapies.

Amperometry approach curve profiling to understand the regulatory mechanisms governing the concentration of intestinal extracellular serotonin.

Enterochromaffin (EC) cells located within the intestinal mucosal epithelium release serotonin (5-HT) to regulate motility tones, barrier function and the immune system. Electroanalytical methodologies have been able to monitor steady state basal extracellular 5-HT levels but are unable to provide insight into how these levels are influenced by key regulatory processes such as release and uptake. We established a new measurement approach, amperometry approach curve profiling, which monitors the extracellular 5-HT level at different electrode-tissue (E-T) distances. Analysis of the current profile can provide information on contributions of regulatory components on the observed extracellular 5-HT level. Measurements were conducted from ex vivo murine ileum and colon using a boron-doped diamond (BDD) microelectrode. Amperometry approach curve profiling coupled with classical pharmacology demonstrated that extracellular 5-HT levels were significantly lower in the colon when compared to the ileum. This difference was due to a greater degree of activity of the 5-HT transporter (SERT) and a reduced amount of 5-HT released from colonic EC cells. The presence of an inhibitory 5-HT4 autoreceptor was observed in the colon, where a 40% increase in extracellular 5-HT was the half maximal inhibitory concentration for activation of the autoreceptor. This novel electroanalytical approach allows estimates of release and re-uptake and their contribution to 5-HT extracellular concentration from intestinal tissue be obtained from a single series of measurements.

Differential CpG methylation at Nnat in the early establishment of beta cell heterogeneity.

AIMS/HYPOTHESIS: Beta cells within the pancreatic islet represent a heterogenous population wherein individual sub-groups of cells make distinct contributions to the overall control of insulin secretion. These include a subpopulation of highly connected 'hub' cells, important for the propagation of intercellular Ca2+ waves. Functional subpopulations have also been demonstrated in human beta cells, with an altered subtype distribution apparent in type 2 diabetes. At present, the molecular mechanisms through which beta cell hierarchy is established are poorly understood. Changes at the level of the epigenome provide one such possibility, which we explore here by focusing on the imprinted gene Nnat (encoding neuronatin [NNAT]), which is required for normal insulin synthesis and secretion. METHODS: Single-cell RNA-seq datasets were examined using Seurat 4.0 and ClusterProfiler running under R. Transgenic mice expressing enhanced GFP under the control of the Nnat enhancer/promoter regions were generated for FACS of beta cells and downstream analysis of CpG methylation by bisulphite sequencing and RNA-seq, respectively. Animals deleted for the de novo methyltransferase DNA methyltransferase 3 alpha (DNMT3A) from the pancreatic progenitor stage were used to explore control of promoter methylation. Proteomics was performed using affinity purification mass spectrometry and Ca2+ dynamics explored by rapid confocal imaging of Cal-520 AM and Cal-590 AM. Insulin secretion was measured using homogeneous time-resolved fluorescence imaging. RESULTS: Nnat mRNA was differentially expressed in a discrete beta cell population in a developmental stage- and DNA methylation (DNMT3A)-dependent manner. Thus, pseudo-time analysis of embryonic datasets demonstrated the early establishment of Nnat-positive and -negative subpopulations during embryogenesis. NNAT expression is also restricted to a subset of beta cells across the human islet that is maintained throughout adult life. NNAT+ beta cells also displayed a discrete transcriptome at adult stages, representing a subpopulation specialised for insulin production, and were diminished in db/db mice. 'Hub' cells were less abundant in the NNAT+ population, consistent with epigenetic control of this functional specialisation. CONCLUSIONS/INTERPRETATION: These findings demonstrate that differential DNA methylation at Nnat represents a novel means through which beta cell heterogeneity is established during development. We therefore hypothesise that changes in methylation at this locus may contribute to a loss of beta cell hierarchy and connectivity, potentially contributing to defective insulin secretion in some forms of diabetes. DATA AVAILABILITY: The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD048465.

Monitoring the electroactive cargo of extracellular vesicles can differentiate various cancer cell lines.

Extracellular vesicles (EVs) are pivotal in cell-to-cell communication due to the array of cargo contained within these vesicles. EVs are considered important biomarkers for identification of disease, however most measurement approaches have focused on monitoring specific surface macromolecular targets. Our study focuses on exploring the electroactive component present within cargo from EVs obtained from various cancer and non-cancer cell lines using a disk carbon fiber microelectrode. Variations in the presence of oxidizable components were observed when the total cargo from EVs were measured, with the highest current detected in EVs from MCF7 cells. There were differences observed in the types of oxidizable species present within EVs from MCF7 and A549 cells. Single entity measurements showed clear spikes due to the detection of oxidizable cargo within EVs from MCF7 and A549 cells. These studies highlight the promise of monitoring EVs through the presence of varying electroactive components within the cargo and can drive a wave of new strategies towards specific detection of EVs for diagnosis and prognosis of various diseases.

Perioperative Considerations in Alpha-Gal Syndrome: A Review.

Galactose-⍺-1, 3-galactose (alpha-gal) is an oligosaccharide found in mammalian tissues that causes allergic reactions in patients with alpha-gal syndrome (AGS). AGS is a hypersensitivity reaction notable for both immediate and delayed allergic and anaphylactic symptoms. As a tick-based disease, AGS has gained increasing prevalence across the United States and can have a significant influence on which medications are safe for patients. Many medications used within the operating room and intensive care units have inactive ingredients that can be mammalian-derived and therefore should be vetted before administering to patients with AGS. Management of patients with AGS involves diligent action in the preoperative and perioperative settings to reduce patient exposure to potentially harmful medications. In conducting a comprehensive risk stratification assessment, the anesthesia team should identify any at-risk patients and determine which medications they have safely tolerated in the past. Despite obtaining a complete history, not all patients with AGS will be identified preoperatively. The perioperative team should understand which common medications pose a risk of containing alpha-gal moieties (e.g., heparins, gelatin capsules, vaccines, lidocaine patches, surgifoam, etc.​​). For this reason, this paper includes a compendium of common anesthetic medications that have been cross-referenced for ingredients that have the potential to cause an AGS reaction. Any potentially unsafe medications have been identified such that medical providers can cross-reference with the ingredients listed at their respective institutions.

Improved Risk-Stratification Scheme for Mismatch-Repair Proficient Stage II Colorectal Cancers Using the Digital Pathology Biomarker QuantCRC.

PURPOSE: There is a need to improve current risk stratification of stage II colorectal cancer to better inform risk of recurrence and guide adjuvant chemotherapy. We sought to examine whether integration of QuantCRC, a digital pathology biomarker utilizing hematoxylin and eosin-stained slides, provides improved risk stratification over current American Society of Clinical Oncology (ASCO) guidelines. EXPERIMENTAL DESIGN: ASCO and QuantCRC-integrated schemes were applied to a cohort of 398 mismatch-repair proficient (MMRP) stage II colorectal cancers from three large academic medical centers. The ASCO stage II scheme was taken from recent guidelines. The QuantCRC-integrated scheme utilized pT3 versus pT4 and a QuantCRC-derived risk classification. Evaluation of recurrence-free survival (RFS) according to these risk schemes was compared using the log-rank test and HR. RESULTS: Integration of QuantCRC provides improved risk stratification compared with the ASCO scheme for stage II MMRP colorectal cancers. The QuantCRC-integrated scheme placed more stage II tumors in the low-risk group compared with the ASCO scheme (62.5% vs. 42.2%) without compromising excellent 3-year RFS. The QuantCRC-integrated scheme provided larger HR for both intermediate-risk (2.27; 95% CI, 1.32-3.91; P = 0.003) and high-risk (3.27; 95% CI, 1.42-7.55; P = 0.006) groups compared with ASCO intermediate-risk (1.58; 95% CI, 0.87-2.87; P = 0.1) and high-risk (2.24; 95% CI, 1.09-4.62; P = 0.03) groups. The QuantCRC-integrated risk groups remained prognostic in the subgroup of patients that did not receive any adjuvant chemotherapy. CONCLUSIONS: Incorporation of QuantCRC into risk stratification provides a powerful predictor of RFS that has potential to guide subsequent treatment and surveillance for stage II MMRP colorectal cancers.

Opportunistic screening for coronary artery calcium deposition using chest radiographs - a multi-objective models with multi-modal data fusion.

Background: To create an opportunistic screening strategy by multitask deep learning methods to stratify prediction for coronary artery calcium (CAC) and associated cardiovascular risk with frontal chest x-rays (CXR) and minimal data from electronic health records (EHR). Methods: In this retrospective study, 2,121 patients with available computed tomography (CT) scans and corresponding CXR images were collected internally (Mayo Enterprise) with calculated CAC scores binned into 3 categories (0, 1-99, and 100+) as ground truths for model training. Results from the internal training were tested on multiple external datasets (domestic (EUH) and foreign (VGHTPE)) with significant racial and ethnic differences and classification performance was compared. Findings: Classification performance between 0, 1-99, and 100+ CAC scores performed moderately on both the internal test and external datasets, reaching average f1-score of 0.66 for Mayo, 0.62 for EUH and 0.61 for VGHTPE. For the clinically relevant binary task of 0 vs 400+ CAC classification, the performance of our model on the internal test and external datasets reached an average AUCROC of 0.84. Interpretation: The fusion model trained on CXR performed better (0.84 average AUROC on internal and external dataset) than existing state-of-the-art models on predicting CAC scores only on internal (0.73 AUROC), with robust performance on external datasets. Thus, our proposed model may be used as a robust, first-pass opportunistic screening method for cardiovascular risk from regular chest radiographs. For community use, trained model and the inference code can be downloaded with an academic open-source license from https://github.com/jeong-jasonji/MTL_CAC_classification . Funding: The study was partially supported by National Institute of Health 1R01HL155410-01A1 award.

Pristine Ti3C2Tx MXene Enables Flexible and Transparent Electrochemical Sensors.

In the era of the internet of things, there exists a pressing need for technologies that meet the stringent demands of wearable, self-powered, and seamlessly integrated devices. Current approaches to developing MXene-based electrochemical sensors involve either rigid or opaque components, limiting their use in niche applications. This study investigates the potential of pristine Ti3C2Tx electrodes for flexible and transparent electrochemical sensing, achieved through an exploration of how material characteristics (flake size, flake orientation, film geometry, and uniformity) impact the electrochemical activity of the outer sphere redox probe ruthenium hexamine using cyclic voltammetry. The optimized electrode made of stacked large Ti3C2Tx flakes demonstrated excellent reproducibility and resistance to bending conditions, suggesting their use for reliable, robust, and flexible sensors. Reducing electrode thickness resulted in an amplified faradaic-to-capacitance signal, which is advantageous for this application. This led to the deposition of transparent thin Ti3C2Tx films, which maintained their best performance up to 73% transparency. These findings underscore its promise for high-performance, tailored sensors, marking a significant stride in advancing MXene utilization in next-generation electrochemical sensing technologies. The results encourage the analytical electrochemistry field to take advantage of the unique properties that pristine Ti3C2Tx electrodes can provide in sensing through more parametric studies.

Evaluating diverse electrode surface patterns of 3D printed carbon thermoplastic electrochemical sensors.

Electrochemical sensing techniques rely on redox reactions taking place at the electrode surface. The configuration of this surface is of the utmost importance in the advancement of electrochemical sensors. The majority of previous electrode manufacturing methods, including 3D printing have produced electrodes with flat surfaces. There is a distinct potential for 3D printing to create intricate and distinctive electrode surface shapes. In the proposed work, 3D printed carbon black polylactic acid electrodes with nine different surface morphologies were made. These were compared to a flat surface electrode. To evaluate the performance of the electrodes, measurements were conducted in three different redox probes (ferrocene methanol, ferricyanide, and dopamine). Our findings highlighted that when electrodes were normalised for the geometric surface area of the electrode, the surface pattern of the electrode surface can impact the observed current and electron transfer kinetics. Electrodes that had a dome and flag pattern on the electrode surface showed the highest oxidation currents and had lower values for the difference between the anodic and cathodic peak current (ΔE). However, designs with rings had lower current values and higher ΔE values. These differences are most likely due to variations in the accessibility of conductive sites on the electrode surface due to the varying surface roughness of different patterned designs. Our findings highlight that when making electrodes using 3D printing, surface patterning of the electrode surface can be used as an effective approach to enhance the performance of the sensor for varying applications.

Liver fibrosis classification from ultrasound using machine learning: a systematic literature review.

PURPOSE: Liver biopsy was considered the gold standard for diagnosing liver fibrosis; however, with advancements in medical technology and increasing awareness of potential complications, the reliance on liver biopsy has diminished. Ultrasound is gaining popularity due to its wider availability and cost-effectiveness. This study examined the machine learning / deep learning (ML/DL) models for non-invasive liver fibrosis classification from ultrasound. METHODS: Following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) protocol, we searched five academic databases using the query. We defined population, intervention, comparison, outcomes, and study design (PICOS) framework for the inclusion. Furthermore, Joana Briggs Institute (JBI) checklist for analytical cross-sectional studies is used for quality assessment. RESULTS: Among the 188 screened studies, 17 studies are selected. The methods are categorized as off-the-shelf (OTS), attention, generative, and ensemble classifiers. Most studies used OTS classifiers that combined pre-trained ML/DL methods with radiomics features to determine fibrosis staging. Although machine learning shows potential for fibrosis classification, there are limited external comparisons of interventions and prospective clinical trials, which limits their applicability. CONCLUSION: With the recent success of ML/DL toward biomedical image analysis, automated solutions using ultrasound are developed for predicting liver diseases. However, their applicability is bounded by the limited and imbalanced retrospective studies having high heterogeneity. This challenge could be addressed by generating a standard protocol for study design by selecting appropriate population, interventions, outcomes, and comparison.

Preoperative and Intraoperative Urine Cultures and Its Association With Postoperative Infection After Ureteroneocystostomy.

OBJECTIVE: To unveil this association, we hypothesize that preoperative and intraoperative urinary tract infection (UTI) will be correlated with postoperative UTI and sepsis occurrence. PATIENTS AND METHODS: The 2020 National Surgical Quality Improvement Program Pediatric (NSQIP-P) data was analyzed for patients undergoing ureteroneocystostomy (UNC) for vesicoureteral reflux (VUR). Patients that underwent UNC for treatment of VUR with urine culture obtained within 2weeks preoperatively or on the day of surgery were identified. The patients were divided into 3 groups: no bacterial growth, bacterial growth with UTI, and bacterial growth polymicrobial growth. Patient demographics and preoperative variables were evaluated. RESULTS: The postoperative urinary tract infection rate of the three groups were 2.0%, 9.2%, and 9.9% for group A, B, C, respectively (P < .001). Postoperative sepsis was noted to be 0.5%, 1.3%, and 3.6% for group A, B, C (P < .01). Additionally, there was a difference between mean operative time (P < .001), mean length of stay (P = .03), and mean days from operation to discharge (P < .01). On adjusted analysis, both groups B and C had higher rates of UTI compared to group A. Group C was also seen to have greater rates of sepsis on adjusted analysis. CONCLUSION: The association found between preoperative UTI with less than 2 species of microorganisms (group B) and postoperative UTI indicates that UTI treatment and antibiotic prophylaxis should be considered when undergoing UNC for VUR. The results of this study may lead to more careful consideration of the use of preoperative and intraoperative urine culture as well as treatment of UTI in pediatric patients with VUR undergoing UNC.

An initiative to assess and improve the resources and patient care processes used among Chest Wall Injury Society collaborative centers study (CWIS-CC2).

BACKGROUND: Over the last two decades, the acute management of rib fractures has changed significantly. In 2021, the Chest Wall injury Society (CWIS) began recognizing centers that epitomize their mission as CWIS Collaborative Centers. The primary aim of this study was to determine the resources, surgical expertise, access to care, and institutional support that are present among centers. METHODS: A survey was performed including all CWIS Collaborative Centers evaluating the resources available at their hospital for the treatment of patients with chest wall injury. Data about each chest wall injury center care process, availability of resources, institutional support, research support, and educational offerings were recorded. RESULTS: Data were collected from 20 trauma centers resulting in an 80% response rate. These trauma centers were made up of 5 international and 15 US-based trauma centers. Eighty percent (16 of 20) have dedicated care team members for the evaluation and management of rib fractures. Twenty-five percent (5 of 20) have a dedicated rib fracture service with a separate call schedule. Staffing for chest wall injury clinics consists of a multidisciplinary team: with attending surgeons in all clinics, 80% (8 of 10) with advanced practice providers and 70% (7 of 10) with care coordinators. Forty percent (8 of 20) of centers have dedicated rib fracture research support, and 35% (7 of 20) have surgical stabilization of rib fracture (SSRF)-related grants. Forty percent (8 of 20) of centers have marketing support, and 30% (8 of 20) have a web page support to bring awareness to their center. At these trauma centers, a median of 4 (1-9) surgeons perform SSRFs. In the majority of trauma centers, the trauma surgeons perform SSRF. CONCLUSION: Considerable similarities and differences exist within these CWIS collaborative centers. These differences in resources are hypothesis generating in determining the optimal chest wall injury center. These findings may generate several patient care and team process questions to optimize patient care, patient experience, provider satisfaction, research productivity, education, and outreach. LEVEL OF EVIDENCE: Therapeutic/Care Management; Level V.

Efficient adversarial debiasing with concept activation vector - Medical image case-studies.

BACKGROUND: A major hurdle for the real time deployment of the AI models is ensuring trustworthiness of these models for the unseen population. More often than not, these complex models are black boxes in which promising results are generated. However, when scrutinized, these models begin to reveal implicit biases during the decision making, particularly for the minority subgroups. METHOD: We develop an efficient adversarial de-biasing approach with partial learning by incorporating the existing concept activation vectors (CAV) methodology, to reduce racial disparities while preserving the performance of the targeted task. CAV is originally a model interpretability technique which we adopted to identify convolution layers responsible for learning race and only fine-tune up to that layer instead of fine-tuning the complete network, limiting the drop in performance RESULTS:: The methodology has been evaluated on two independent medical image case-studies - chest X-ray and mammograms, and we also performed external validation on a different racial population. On the external datasets for the chest X-ray use-case, debiased models (averaged AUC 0.87 ) outperformed the baseline convolution models (averaged AUC 0.57 ) as well as the models trained with the popular fine-tuning strategy (averaged AUC 0.81). Moreover, the mammogram models is debiased using a single dataset (white, black and Asian) and improved the performance on an external datasets (averaged AUC 0.8 to 0.86 ) with completely different population (primarily Hispanic patients). CONCLUSION: In this study, we demonstrated that the adversarial models trained only with internal data performed equally or often outperformed the standard fine-tuning strategy with data from an external setting. The adversarial training approach described can be applied regardless of predictor's model architecture, as long as the convolution model is trained using a gradient-based method. We release the training code with academic open-source license - https://github.com/ramon349/JBI2023_TCAV_debiasing.

Differential CpG methylation at Nnat in the early establishment of beta cell heterogeneity.

Aims/hypothesis: Beta cells within the pancreatic islet represent a heterogenous population wherein individual sub-groups of cells make distinct contributions to the overall control of insulin secretion. These include a subpopulation of highly-connected 'hub' cells, important for the propagation of intercellular Ca2+ waves. Functional subpopulations have also been demonstrated in human beta cells, with an altered subtype distribution apparent in type 2 diabetes. At present, the molecular mechanisms through which beta cell hierarchy is established are poorly understood. Changes at the level of the epigenome provide one such possibility which we explore here by focussing on the imprinted gene neuronatin (Nnat), which is required for normal insulin synthesis and secretion. Methods: Single cell RNA-seq datasets were examined using Seurat 4.0 and ClusterProfiler running under R. Transgenic mice expressing eGFP under the control of the Nnat enhancer/promoter regions were generated for fluorescence-activated cell (FAC) sorting of beta cells and downstream analysis of CpG methylation by bisulphite and RNA sequencing, respectively. Animals deleted for the de novo methyltransferase, DNMT3A from the pancreatic progenitor stage were used to explore control of promoter methylation. Proteomics was performed using affinity purification mass spectrometry and Ca2+ dynamics explored by rapid confocal imaging of Cal-520 and Cal-590. Insulin secretion was measured using Homogeneous Time Resolved Fluorescence Imaging. Results: Nnat mRNA was differentially expressed in a discrete beta cell population in a developmental stage- and DNA methylation (DNMT3A)-dependent manner. Thus, pseudo-time analysis of embryonic data sets demonstrated the early establishment of Nnat-positive and negative subpopulations during embryogenesis. NNAT expression is also restricted to a subset of beta cells across the human islet that is maintained throughout adult life. NNAT+ beta cells also displayed a discrete transcriptome at adult stages, representing a sub-population specialised for insulin production, reminiscent of recently-described "ßHI" cells and were diminished in db/db mice. 'Hub' cells were less abundant in the NNAT+ population, consistent with epigenetic control of this functional specialization. Conclusions/interpretation: These findings demonstrate that differential DNA methylation at Nnat represents a novel means through which beta cell heterogeneity is established during development. We therefore hypothesise that changes in methylation at this locus may thus contribute to a loss of beta cell hierarchy and connectivity, potentially contributing to defective insulin secretion in some forms of diabetes.

Opportunistic assessment of ischemic heart disease risk using abdominopelvic computed tomography and medical record data: a multimodal explainable artificial intelligence approach.

Current risk scores using clinical risk factors for predicting ischemic heart disease (IHD) events-the leading cause of global mortality-have known limitations and may be improved by imaging biomarkers. While body composition (BC) imaging biomarkers derived from abdominopelvic computed tomography (CT) correlate with IHD risk, they are impractical to measure manually. Here, in a retrospective cohort of 8139 contrast-enhanced abdominopelvic CT examinations undergoing up to 5 years of follow-up, we developed multimodal opportunistic risk assessment models for IHD by automatically extracting BC features from abdominal CT images and integrating these with features from each patient's electronic medical record (EMR). Our predictive methods match and, in some cases, outperform clinical risk scores currently used in IHD risk assessment. We provide clinical interpretability of our model using a new method of determining tissue-level contributions from CT along with weightings of EMR features contributing to IHD risk. We conclude that such a multimodal approach, which automatically integrates BC biomarkers and EMR data, can enhance IHD risk assessment and aid primary prevention efforts for IHD. To further promote research, we release the Opportunistic L3 Ischemic heart disease (OL3I) dataset, the first public multimodal dataset for opportunistic CT prediction of IHD.

Challenges and solutions of echocardiography generalization for deep learning: a study in patients with constrictive pericarditis.

Purpose: The inherent characteristics of transthoracic echocardiography (TTE) images such as low signal-to-noise ratio and acquisition variations can limit the direct use of TTE images in the development and generalization of deep learning models. As such, we propose an innovative automated framework to address the common challenges in the process of echocardiography deep learning model generalization on the challenging task of constrictive pericarditis (CP) and cardiac amyloidosis (CA) differentiation. Approach: Patients with a confirmed diagnosis of CP or CA and normal cases from Mayo Clinic Rochester and Arizona were identified to extract baseline demographics and the apical 4 chamber view from TTE studies. We proposed an innovative preprocessing and image generalization framework to process the images for training the ResNet50, ResNeXt101, and EfficientNetB2 models. Ablation studies were conducted to justify the effect of each proposed processing step in the final classification performance. Results: The models were initially trained and validated on 720 unique TTE studies from Mayo Rochester and further validated on 225 studies from Mayo Arizona. With our proposed generalization framework, EfficientNetB2 generalized the best with an average area under the curve (AUC) of 0.96 (±0.01) and 0.83 (±0.03) on the Rochester and Arizona test sets, respectively. Conclusions: Leveraging the proposed generalization techniques, we successfully developed an echocardiography-based deep learning model that can accurately differentiate CP from CA and normal cases and applied the model to images from two sites. The proposed framework can be further extended for the development of echocardiography-based deep learning models.

Association of Coronary Artery Calcium Detected by Routine Ungated CT Imaging With Cardiovascular Outcomes.

BACKGROUND: Coronary artery calcium (CAC) is a strong predictor of cardiovascular events across all racial and ethnic groups. CAC can be quantified on nonelectrocardiography (ECG)-gated computed tomography (CT) performed for other reasons, allowing for opportunistic screening for subclinical atherosclerosis. OBJECTIVES: The authors investigated whether incidental CAC quantified on routine non-ECG-gated CTs using a deep-learning (DL) algorithm provided cardiovascular risk stratification beyond traditional risk prediction methods. METHODS: Incidental CAC was quantified using a DL algorithm (DL-CAC) on non-ECG-gated chest CTs performed for routine care in all settings at a large academic medical center from 2014 to 2019. We measured the association between DL-CAC (0, 1-99, or ≥100) with all-cause death (primary outcome), and the secondary composite outcomes of death/myocardial infarction (MI)/stroke and death/MI/stroke/revascularization using Cox regression. We adjusted for age, sex, race, ethnicity, comorbidities, systolic blood pressure, lipid levels, smoking status, and antihypertensive use. Ten-year atherosclerotic cardiovascular disease risk was calculated using the pooled cohort equations. RESULTS: Of 5,678 adults without ASCVD (51% women, 18% Asian, 13% Hispanic/Latinx), 52% had DL-CAC >0. Those with DL-CAC ≥100 had an average 10-year ASCVD risk of 24%; yet, only 26% were on statins. After adjustment, patients with DL-CAC ≥100 had increased risk of death (HR: 1.51; 95% CI: 1.28-1.79), death/MI/stroke (HR: 1.57; 95% CI: 1.33-1.84), and death/MI/stroke/revascularization (HR: 1.69; 95% CI: 1.45-1.98) compared with DL-CAC = 0. CONCLUSIONS: Incidental CAC ≥100 was associated with an increased risk of all-cause death and adverse cardiovascular outcomes, beyond traditional risk factors. DL-CAC from routine non-ECG-gated CTs identifies patients at increased cardiovascular risk and holds promise as a tool for opportunistic screening to facilitate earlier intervention.

Contrastive diagnostic embedding (CDE) model for automated coding - A case study using emergency department encounters.

BACKGROUND: Billing codes are utilized for medical reimbursement, clinical quality metric valuation and for epidemiologic purposes to report and follow disease trends and outcomes. The current paradigm of manual coding can be expensive, time-consuming, and subject to human error. Though automation of the billing codes has been widely reported in the literature via rule-based and supervised approaches, existing strategies lack generalizability and robustness towards large and constantly changing ICD hierarchical structure. METHOD: We propose a weakly supervised training strategy by leveraging contrastive learning, contrastive diagnosis embedding (CDE) to capture the fine semantic variations between the diagnosis codes. The approach consists of a two-phase contrastive training for generating the semantic embedding space adapted to incorporate hierarchical information of ICD-10 vocabulary and a weakly supervised retrieval scheme. Core strength of the proposed method is that it puts no limit on the 70 K ICD-10 codes set and can handle all rare codes for coding the diagnosis. RESULTS: Our CDE model outperformed string-based partial matching and ClinicalBERT embedding on three test cases (a retrospective testset, a prospective testset, and external testset) and produced an accurate prediction of rare and newly introduced diagnosis codes. A detailed ablation study showed the importance of each phase of the proposed multi-phase training. Each successive phase of training - ICD-10 group sensitive training (phase 1.1), ICD-10 subgroup sensitive training (phase 1.2), free-text diagnosis description-based training (phase 2) - improved performance beyond the previous phase of training. The model also outperformed existing supervised models like CAML and PLM-ICD and produced satisfactory performance on the rare codes. CONCLUSION: Compared to the existing rule-based and supervised models, the proposed weakly supervised contrastive learning overcomes the limitations in terms of generalization capability and increases the robustness of the automated billing. Such a model will allow flexibility through accurate billing code automation for practice convergence and gains efficiencies in a value-based care payment environment.

Synthetic dual-energy CT reconstruction from single-energy CT Using artificial intelligence.

PURPOSE: To develop and assess the utility of synthetic dual-energy CT (sDECT) images generated from single-energy CT (SECT) using two state-of-the-art generative adversarial network (GAN) architectures for artificial intelligence-based image translation. METHODS: In this retrospective study, 734 patients (389F; 62.8 years ± 14.9) who underwent enhanced DECT of the chest, abdomen, and pelvis between January 2018 and June 2019 were included. Using 70-keV as the input images (n = 141,009) and 50-keV, iodine, and virtual unenhanced (VUE) images as outputs, separate models were trained using Pix2PixHD and CycleGAN. Model performance on the test set (n = 17,839) was evaluated using mean squared error, structural similarity index, and peak signal-to-noise ratio. To objectively test the utility of these models, synthetic iodine material density and 50-keV images were generated from SECT images of 16 patients with gastrointestinal bleeding performed at another institution. The conspicuity of gastrointestinal bleeding using sDECT was compared to portal venous phase SECT. Synthetic VUE images were generated from 37 patients who underwent a CT urogram at another institution and model performance was compared to true unenhanced images. RESULTS: sDECT from both Pix2PixHD and CycleGAN were qualitatively indistinguishable from true DECT by a board-certified radiologist (avg accuracy 64.5%). Pix2PixHD had better quantitative performance compared to CycleGAN (e.g., structural similarity index for iodine: 87% vs. 46%, p-value < 0.001). sDECT using Pix2PixHD showed increased bleeding conspicuity for gastrointestinal bleeding and better removal of iodine on synthetic VUE compared to CycleGAN. CONCLUSIONS: sDECT from SECT using Pix2PixHD may afford some of the advantages of DECT.