Reliable prognostic biomarkers are needed to guide treatment decisions for patients with metastatic kidney cancer receiving immune checkpoint inhibitors (ICIs). Radiomics biomarkers leveraging quantitative imaging features from longitudinal CT scans have demonstrated prognostic utility across multiple tumor types, including advanced non-small cell lung cancer (NSCLC). Evaluating such biomarkers in additional tumor types may further enhance personalized treatment approaches. Evaluate the performance of a deep learning biomarker for predicting overall survival (OS) in metastatic kidney cancer patients receiving ICIs. Serial computed tomography response score (Serial CTRS) is a fully automated deep learning radiomics biomarker that predicts OS by analyzing paired baseline and early-treatment thoracic CT scans, typically capturing thoracic and upper abdominal disease. Serial CTRS was previously validated in advanced NSCLC patients receiving programmed death-ligand 1 (PD-L1) ICIs, demonstrating superior OS prediction compared to conventional RECIST and tumor volume metrics in retrospective real-world and clinical trial datasets. This study retrospectively analyzed paired baseline (within 90 days prior to ICI initiation; median: 21 days prior) and follow-up (28–120 days post-ICI initiation; median: 80 days) thoracic CT scans from 117 metastatic kidney cancer patients treated with ICIs within Providence Health System. Of these, 87 patients (median age: 66 years; IQR: 59–72) with available paired scans were included. Predictive performance of Serial CTRS for OS was assessed using Cox proportional hazards models, concordance index (C-index), and area under the receiver operating characteristic curve (ROC-AUC) for OS at 6, 12, and 24 months. Serial CTRS demonstrated significant association with OS, yielding robust risk stratification (C-index: 0.73; 95% CI: 0.65–0.80). ROC-AUC values showed strong predictive accuracy for OS at 6 months (0.87; 95% CI: 0.79–0.94), 12 months (0.78; 95% CI: 0.66–0.90), and 24 months (0.73; 95% CI: 0.60–0.87). Kaplan-Meier analysis using predetermined thresholds from prior NSCLC datasets revealed clear survival stratification among Serial CTRS groups: low- versus high-survival probability (HR=5.01; 95% CI: 2.15–11.68), low- versus intermediate-survival probability (HR=3.25; 95% CI: 1.58–6.69), and intermediate- versus high-survival probability (HR=1.84; 95% CI: 0.82–4.15). Serial CTRS demonstrated robust and significant predictive utility for OS in metastatic kidney cancer patients receiving ICIs, consistent with previous validation in advanced NSCLC. Its fully automated methodology, which requires no manual lesion annotations, may facilitate scalable and objective clinical implementation enhancing prognostication and optimizing treatment stratification in oncology clinical trials and clinical practice. Further prospective validation exploring integration of Serial CTRS into clinical trial designs is warranted. Chiharu Sako, Taly G. Schmidt, Beatriz G. Lourenco, Karishma Sewaramani, Ross McCall, Ryan Beasley, Arpan A. Patel, Dwight H. Owen, Arya Amini, Ronan J. Kelly, Ray D. Page, Jean-Paul Beregi, Stephane Sanchez, Olivier Gevaert, George R. Simon, Ravi B. Parikh, Petr Jordan, Brendan D. Curti. Validation of a Deep Learning Serial Computed Tomography Response Biomarker for Predicting Overall Survival in Metastatic Kidney Cancer Treated with Immune Checkpoint Inhibitors [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Artificial Intelligence and Machine Learning; 2025 Jul 10-12; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(13_Suppl):Abstract nr A029.

e13689 Background: Machine learning models using radiomic features have demonstrated predictive and prognostic value in oncology applications. One challenge in translating these models into clinical use is the wide variation in imaging protocols across clinical sites. Variations in reconstruction kernels in CT images can alter radiomic features, introducing bias and impacting model generalizability. We investigated the effects of a physics-based post-processing harmonization method on (1) pre-defined radiomic features, and (2) the performance of a deep learning model to predict immune checkpoint inhibitor (ICI) response in advanced NSCLC. Methods: This study was performed on an internally curated, multi-institutional real-world dataset of chest CT scans from 1,188 advanced NSCLC patients treated with PD-(L)1 ICIs in academic and community settings from the US and Europe. The dataset included images spanning eight manufacturers, over 100 scanner models, and 31 reconstruction kernels. A post-processing harmonization algorithm, leveraging phantom-derived protocol libraries and an online noise power spectrum estimation method, was used to standardize images to a common kernel, correcting for reconstruction variability. In the first investigation, 85 radiomic features (e.g., shape, texture) were extracted from the native images and from images harmonized to a common kernel and compared. Performance of the deep learning model in predicting ICI therapy response was quantified by the area under the ROC curve for Progression-Free Survival at 6 months (PFS6 AUC) and Overall Survival at 12 months (OS12 AUC), comparing models trained with and without harmonization. The models were evaluated on an independent test set of 92 patients who received ICI in the first line as a monotherapy from an institution not used for training. Results: Variability in radiomic features due to convolution kernels was significant; the mean difference between images reconstructed by the Lung kernel compared to the Standard kernel was 60%, with a maximum of 359%. After Lung kernel images were harmonized to match the appearance of the Standard images, the mean difference was reduced to 6%, with a maximum of 23%. The deep learning model performance improved when trained on harmonized images, increasing PFS6 AUC from 0.64 to 0.70 and OS12 AUC from 0.75 to 0.79. Conclusions: The large variation in reconstruction kernels used in clinical CT imaging practice may confound radiomic features and compromise the performance of predictive models. Post-processing image harmonization successfully reduced these variations, reducing the bias in radiomic features and improving the predictive performance of a deep learning model for ICI therapy response. This study demonstrates the importance of harmonization techniques to ensure imaging biomarkers that are generalizable.

Early response assessment and identification of patients (pts) with NSCLC likely to derive long-term benefit from immune checkpoint inhibitors (ICIs) are crucial for treatment planning and drug development. We developed a deep learning pipeline using a large multi-institute real-world dataset to generate a serial CT response score (SerialCTRS) from paired pre-treatment and 12wk scans, estimating probability of 1y OS. External validation applied SerialCTRS to an expansion cohort of a single-arm phase 1 clinical trial of dostarlimab as ≥2L therapy for advanced NSCLC (GARNET [cohort E], NCT02715284). OS hazard ratios (HRs) of SerialCTRS were compared to tumor volume change derived from manual segmentations and 12 wk RECIST 1.1 overall response using the same input scans. Of 67 clinical trial pts, 49 were evaluable for all 3 models predicting OS. Matching group sizes to RECIST overall response, 13 pts (26%, the number with RECIST PD) had low probability of 1y OS (SerialCTRS 0.12-0.64), 21 pts (43%, number with SD) had medium probability (0.64-0.82), and 15 pts (31%, analogous to CR/PR) had high probability of 1y OS (0.83-0.90). HRs of adjacent categories predicting OS showed superior performance of SerialCTRS, especially for SD/medium vs CR/PR/high probability (HR 2.68 [95% CI 1.09-6.59]; Table). Continuous SerialCTRS predicted OS with a concordance index of 0.717 (0.733 in subset with RECIST 12 wk SD) and remained a significant predictor of OS in bivariable survival models controlling for known predictors of OS such as age, ECOG, stage at diagnosis (all n=54), and best overall response from previous therapy (n=49). Deep learning-based SerialCTRS, without manual annotation, improved prediction of OS vs RECIST and tumor volume change in the NSCLC cohort of the GARNET trial. Brenda F. Kurland, Chiharu Sako, Jing He, Marius de Groot, Taly G. Schmidt, Arpan A. Patel, Dwight H. Owen, Arya Amini, Brendan D. Curti, Ronan J. Kelly, Ray D. Page, Aurelie Swalduz, Jean-Paul Beregi, Jan Chrusciel, Stephane Sanchez, Richard Rosenberg, Jakob Weiss, An Liu, Olivier Gevaert, George R. Simon, Ravi B. Parikh, Alma Hart, Petr Jordan, Jasper van der Aart. Deep learning response score using baseline and 12-week RECIST chest CTs enhances overall survival (OS) prediction in advanced NSCLC: external validation in a trial with dostarlimab [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 736.

1559 Background: Identifying advanced non–small cell lung cancer (aNSCLC) patients who derive long-term benefit from immune checkpoint inhibitors (ICIs) remains a significant challenge. Radiomic analyses, particularly leveraging deep learning, hold promise for improving prognostic accuracy beyond tumor size metrics. We developed serialCTRS, a novel biomarker using deep learning to quantify thoracic CT changes from baseline to 3 months post-treatment, predicting overall survival (OS) in patients receiving PD-(L)1 inhibitors. Methods: SerialCTRS was previously trained and validated on a multi-institutional Real-World Dataset (RWD) (training: 1,171 aNSCLC patients, 14,424 CT scans; validation: 612 patients; Sako et al. SITC, 2024). For this study, we retrospectively validated serialCTRS in two distinct cohorts of aNSCLC patients: (1) a clinical trial (N = 52) treated with the PD-1 inhibitor sasanlimab in the second or later line and (2) a fully blinded RWD from Baylor Scott & White Health system (N = 147), an institution not used for training. The pipeline—spanning image quality control, preprocessing, feature extraction, and survival modeling—operated without manual annotations. To enhance interpretability, we developed 3D submodels for prognostic signals related to (i) tumor burden, (ii) body composition, and (iii) lung vasculature. Predictive performance was compared to RECIST 1.1 using concordance index (c-index) and ROC-AUC for 24-month OS (OS24 AUC). Results: SerialCTRS outperformed RECIST in OS prediction and remained a significant predictor after multivariate adjustments with other known predictors including age, sex, PD-L1 TPS, and NLR across both validation cohorts. In the sasanlimab cohort, serialCTRS achieved a c-index of 0.77, surpassing RECIST (0.72), with an OS24 AUC of 0.86 (95% CI: 0.74–0.98). In the Baylor cohort, serialCTRS demonstrated a c-index of 0.68 vs. RECIST (0.62) and an OS24 AUC of 0.76 (0.67–0.86). Submodels targeting individual components achieved c-indices of 0.65 (tumor burden), 0.61 (body composition), and 0.61 (vasculature) in the sasanlimab cohort, and 0.63, 0.61, and 0.59, respectively, in the Baylor cohort. Combining the submodels improved c-indices to 0.69 (sasanlimab) and 0.66 (Baylor), demonstrating complementary signal among radiographic features. Conclusions: SerialCTRS outperformed RECIST 1.1 in predicting OS in independent clinical trial and RWD datasets. Interpretable submodels highlighted the prognostic value of tumor burden, body composition, and vasculature changes. SerialCTRS offers a promising tool for personalizing therapy and accelerating drug development in aNSCLC, with a fully automated pipeline for robust and scalable clinical use. Future work will focus on larger, more diverse cohorts to validate utility in guiding precision oncology.

PURPOSEThis study developed and validated a novel deep learning radiomic biomarker to estimate response to immune checkpoint inhibitor (ICI) therapy in advanced non-small cell lung cancer (NSCLC) using real-world data (RWD) and clinical trial data.MATERIALS AND METHODSRetrospective RWD of 1,829 patients with advanced NSCLC treated with PD-(L)1 ICIs were collected from 10 academic and community institutions in the United States and Europe. The RWD included data sets for discovery (Data Set A-Discovery, n = 1,173) and independent test (Data Set B, n = 458). A radiomic pipeline, containing a deep learning feature extractor and a survival model, generated the computed tomography (CT) response score (CTRS) applied to the pretreatment routine CT/positron emission tomography (PET)-CT scan. An enhanced CTRS (eCTRS) also incorporated age, sex, treatment line, and lesion annotations. Performance was evaluated against progression-free survival (PFS) and overall survival (OS). Biomarker generalizability was further evaluated using a secondary analysis of a prospective clinical trial (ClinicalTrials.gov identifier: NCT02573259) evaluating the PD-1 inhibitor sasanlimab in second or later line of treatment (Data Set C, n = 54).RESULTSIn RWD Test Data Set B, the CTRS identified patients with a high probability of response to ICI with a PFS hazard ratio (HR) of 0.46 (95% CI, 0.26 to 0.82) and an OS HR of 0.50 (95% CI, 0.28 to 0.92) in the first-line ICI monotherapy cohort, after adjustment for baseline covariates including the PD-L1 tumor proportion score. In Clinical Trial Data Set C, the CTRS demonstrated an adjusted PFS HR of 1.03 (95% CI, 0.43 to 2.47) and an OS HR of 0.33 (95% CI, 0.14 to 0.91). The CTRS and eCTRS outperformed traditional imaging biomarkers of lesion size in PFS and OS for RWD Test Data Set B and in OS for the Clinical Trial Data Set.CONCLUSIONThe study developed and validated a deep learning radiomic biomarker using pretreatment routine CT/PET-CT scans to identify ICI benefit in advanced NSCLC.

Background: Photon-counting CT (PCCT) systems acquire multiple spectral measurements at high spatial resolution, providing numerous image quality benefits while also increasing the amount of data that must be transferred through the gantry slip ring. Purpose: This study proposes a lossy method to compress photon-counting CT data using eigenvector analysis, with the goal of providing image quality sufficient for applications that require a rapid initial reconstruction, such as to confirm anatomical coverage, scan quality, and to support automated advanced applications. The eigenbin compression method was experimentally evaluated on a clinical silicon PCCT prototype system. Methods: The proposed eigenbin method performs principal component analysis (PCA) on a set of PCCT calibration measurements. PCA finds the orthogonal axes or eigenvectors, which capture the maximum variance in the N dimensional photon-count data space, where N is the number of acquired energy bins. To reduce the dimensionality of the PCCT data, the data are linearly transformed into a lower dimensional space spanned by the M < N eigenvectors with highest eigenvalues (i.e., the vectors that account for most of the information in the data). Only M coefficients are then transferred per measurement, which we term eigenbin values. After transmission, the original N energy-bin measurements are estimated as a linear combination of the M eigenvectors. Two versions of the eigenbin method were investigated: pixel-specific and pixel-general. The pixel-specific eigenbin method determines eigenvectors for each individual detector pixel, while the more practically realizable pixel-general eigenbin method finds one set of eigenvectors for the entire detector array. The eigenbin method was experimentally evaluated by scanning a 20 cm diameter Gammex Multienergy phantom with different material inserts on a clinical silicon-based PCCT prototype. The method was evaluated with the number of eigenbins varied between two and four. In each case, the eigenbins were used to estimate the original 8-bin data, after which material decomposition was performed. The mean, standard deviation, and contrast-to-noise ratio (CNR) of values in the reconstructed basis and virtual monoenergetic images (VMI) were compared for the original 8-bin data and for the eigenbin data. Results: The pixel-specific eigenbin method reduced photon-counting CT data size by a factor of four with <5% change in mean values and a small noise penalty (mean change in noise of <12%, maximum change in noise of 20% for basis images). The pixel-general eigenbin compression method reduced data size by a factor of 2.67 with <5% change in mean values and a less than 10% noise penalty in the basis images (average noise penalty <= 5%). The noise penalty and errors were less for the VMIs than for the basis images, resulting in <5% change in CNR in the VMIs. Conclusion: The eigenbin compression method reduced photon-counting CT data size by a factor of two to four with less than 5% change in mean values, noise penalty of less than 10%-20%, and change in CNR ranging from 15% decrease to 24% increase. Eigenbin compression reduces the data transfer time and storage space of photon-counting CT data for applications that require rapid initial reconstructions.

Organ segmentation from CT images is critical in the early diagnosis of diseases, progress monitoring, pre-operative planning, radiation therapy planning, and CT dose estimation. However, data limitation remains one of the main challenges in medical image segmentation tasks. This challenge is particularly huge in pediatric CT segmentation due to children’s heightened sensitivity to radiation. In order to address this issue, we propose a novel segmentation framework with a built-in auxiliary classifier generative adversarial network (ACGAN) that conditions age, simultaneously generating additional features during training. The proposed conditional feature generation segmentation network (CFG-SegNet) was trained on a single loss function and used 2.5D segmentation batches. Our experiment was performed on a dataset with 359 subjects (180 male and 179 female) aged from 5 days to 16 years and a mean age of 7 years. CFG-SegNet achieved an average segmentation accuracy of 0.681 dice similarity coefficient (DSC) on the prostate, 0.619 DSC on the uterus, 0.912 DSC on the liver, and 0.832 DSC on the heart with four-fold cross-validation. We compared the segmentation accuracy of our proposed method with previously published U-Net results, and our network improved the segmentation accuracy by 2.7%, 2.6%, 2.8%, and 3.4% for the prostate, uterus, liver, and heart, respectively. The results indicate that our high-performing segmentation framework can more precisely segment organs when limited training images are available.

BACKGROUND The constrained one-step spectral CT Image Reconstruction method (cOSSCIR) has been developed to estimate basis material maps directly from spectral CT data using a model of the polyenergetic x-ray transmissions and incorporating convex constraints into the inversion problem. This 'one-step' approach has been shown to stabilize the inversion in the case of photon-counting CT, and may provide similar benefits to dual-kV systems that utilize integrating detectors. Since the approach does not require the same rays be acquired for every spectral measurement, cOSSCIR can apply to dual energy protocols and systems used clinically, such as fast and slow kV switching systems and dual source scanning. PURPOSE The purpose of this study is to investigate the use of cOSSCIR applied to dual-kV data, using both registered and unregistered spectral acquisitions, specifically slow and fast kV switching imaging protocols. For this application, cOSSCIR is investigated using inverse crime simulations and dual-kV experiments. This study is the first demonstration of cOSSCIR on the dual-kV reconstruction problem. METHODS An integrating detector model was developed for the purpose of reconstructing dual-kV data, and an inverse crime study was used to validate the detector model within the cOSSCIR framework using a simulated pelvic phantom. Experiments were also used to evaluate cOSSCIR on the dual energy problem. Dual-kV data was obtained from a physical phantom containing analogs of adipose, bone, and liver tissues, with the aim of recovering the material coefficients in the bone and adipose basis material maps. cOSSCIR was applied to acquisitions where all rays performed both spectral measurements (registered) and fast and slow kV switching acquisitions (unregistered). cOSSCIR was also compared to two image-domain decomposition approaches, where image-domain methods are the conventional approach for decomposing unregistered spectral data. RESULTS Simulations demonstrate the application of cOSSCIR to the dual-kV inversion problem by successfully recovering the material basis maps on ideal data, while further showing that unregistered data presents a more challenging inversion problem. In our experimental reconstructions, the recovered basis material coefficient errors were found to be less than 6.5% in the bone, adipose, and liver regions for both registered and unregistered protocols. Similarly, the errors were less than 4% in the 50 keV virtual mono-energetic images, and the recovered material decomposition vectors nearly overlap their corresponding ground-truth vectors. Additionally, a preliminary two material decomposition study of iodine quantification recovered an average concentration of 9.2 mg/mL from a 10 mg/mL experimental iodine analog. CONCLUSIONS Using our integrating detector and spectral models, cOSCCIR is capable of accurately recovering material basis maps from dual-kV data for both registered and unregistered data. The material decomposition quantification compare favorably to the image domain approaches, and our results were not affected by the imaging protocol. Our results also suggest the extension of cOSSCIR to iodine quantification using two material decomposition.

Photon counting detectors provide improved resolution and dose efficiency compared to scintillating detectors, with the potential for improved material decomposition. However, the measured counts may be inaccurate due to pulse pileup, which can cause material decomposition errors. Prior work demonstrated that Neural Networks (NN) can perform accurate material decomposition for a range of flux levels when trained at each specific tube current [1]. However, training a NN for each tube current is impractical for diagnostic CT because the tube current is continuously modulated. This study investigates a material decomposition NN trained and applied across a range of tube current settings. The NN was trained using calibration step-wedge data from a range of tube currents representing flux levels of 14% to 51.3% of the maximum detector count rate. We refer to this network as ‘flux-independent’ as it can be applied to flux levels not seen in training. The material decomposition accuracy of the flux-independent network was evaluated and compared to that of a NN trained at one flux level, (i.e., ‘flux-specific’ network). The networks were first compared using data from an experimentally-verified simulation model, while experimental evaluation is underway.

Metal artifacts in CT reconstructions can negatively a_ect the diagnostic utility of CT imaging, as the degradation of image quality is generally not localized. One approach to addressing metal artifacts involves incorporating accurate models of the x-ray transmission into the inversion procedure. In the constrained one-step spectral CT image reconstruction method (cOSSCIR), the basis material maps are computed directly from spectral transmission measurements using a non-linear spectral model. When applied to the photon-counting problem, cOSSCIR is capable of removing artifacts due to beam hardening, noise, and photon starvation solely using an accurate model of the x-ray spectrum and detector response. In this study we investigate reconstructing experimental dual-kV data corrupted by metal, and present approaches for reducing artifacts within the framework of cOSSCIR.

Positive margin status after breast-conserving surgery (BCS) is a predictor of higher rates of local recurrence. Intraoperative margin assessment aims to achieve negative surgical margin status at the first operation, thus reducing the re-excision rates that are usually associated with potential surgical complications, increased medical costs, and mental pressure on patients. Microscopy with ultraviolet surface excitation (MUSE) can rapidly image tissue surfaces with subcellular resolution and sharp contrasts by utilizing the nature of the thin optical sectioning thickness of deep ultraviolet light. We have previously imaged 66 fresh human breast specimens that were topically stained with propidium iodide and eosin Y using a customized MUSE system. To achieve objective and automated assessment of MUSE images, a machine learning model is developed for binary (tumor vs. normal) classification of obtained MUSE images. Features extracted by texture analysis and pre-trained convolutional neural networks (CNN) have been investigated for sample descriptions. A sensitivity, specificity, and accuracy better than 90% have been achieved for detecting tumorous specimens. The result suggests the potential of MUSE with machine learning being utilized for intraoperative margin assessment during BCS.

BACKGROUND Spectral CT material decomposition provides quantitative information but is challenged by the instability of the inversion into basis materials. We have previously proposed the constrained One-Step Spectral CT Image Reconstruction (cOSSCIR) algorithm to stabilize the material decomposition inversion by directly estimating basis material images from spectral CT data. cOSSCIR was previously investigated on phantom data. PURPOSE This study investigates the performance of cOSSCIR using head CT datasets acquired on a clinical photon-counting CT (PCCT) prototype. This is the first investigation of cOSSCIR for large-scale, anatomically complex, clinical PCCT data. The cOSSCIR decomposition is preceded by a spectrum estimation and nonlinear counts correction calibration step to address nonideal detector effects. METHODS Head CT data were acquired on an early prototype clinical PCCT system using an edge-on silicon detector with eight energy bins. Calibration data of a step wedge phantom were also acquired and used to train a spectral model to account for the source spectrum and detector spectral response, and also to train a nonlinear counts correction model to account for pulse pileup effects. The cOSSCIR algorithm optimized the bone and adipose basis images directly from the photon counts data, while placing a grouped total variation (TV) constraint on the basis images. For comparison, basis images were also reconstructed by a two-step projection-domain approach of Maximum Likelihood Estimation (MLE) for decomposing basis sinograms, followed by filtered backprojection (MLE + FBP) or a TV minimization algorithm (MLE + TVmin ) to reconstruct basis images. We hypothesize that the cOSSCIR approach will provide a more stable inversion into basis images compared to two-step approaches. To investigate this hypothesis, the noise standard deviation in bone and soft-tissue regions of interest (ROIs) in the reconstructed images were compared between cOSSCIR and the two-step methods for a range of regularization constraint settings. RESULTS cOSSCIR reduced the noise standard deviation in the basis images by a factor of two to six compared to that of MLE + TVmin , when both algorithms were constrained to produce images with the same TV. The cOSSCIR images demonstrated qualitatively improved spatial resolution and depiction of fine anatomical detail. The MLE + TVmin algorithm resulted in lower noise standard deviation than cOSSCIR for the virtual monoenergetic images (VMIs) at higher energy levels and constraint settings, while the cOSSCIR VMIs resulted in lower noise standard deviation at lower energy levels and overall higher qualitative spatial resolution. There were no statistically significant differences in the mean values within the bone region of images reconstructed by the studied algorithms. There were statistically significant differences in the mean values within the soft-tissue region of the reconstructed images, with cOSSCIR producing mean values closer to the expected values. CONCLUSIONS The cOSSCIR algorithm, combined with our previously proposed spectral model estimation and nonlinear counts correction method, successfully estimated bone and adipose basis images from high resolution, large-scale patient data from a clinical PCCT prototype. The cOSSCIR basis images were able to depict fine anatomical details with a factor of two to six reduction in noise standard deviation compared to that of the MLE + TVmin two-step approach.

Purpose Organ autosegmentation efforts to date have largely been focused on adult populations, due to limited availability of pediatric training data. Pediatric patients may present additional challenges for organ segmentation. This paper describes a dataset of 359 pediatric chest-abdomen-pelvis and abdomen-pelvis Computed Tomography (CT) images with expert contours of up to 29 anatomical organ structures to aid in the evaluation and development of autosegmentation algorithms for pediatric CT imaging. Acquisition and validation methods The dataset collection consists of axial CT images in Digital Imaging and Communications in Medicine (DICOM) format of 180 male and 179 female pediatric chest-abdomen-pelvis or abdomen-pelvis exams acquired from one of three CT scanners at Children's Wisconsin. The datasets represent random pediatric cases based upon routine clinical indications. Subjects ranged in age from 5 days to 16 years, with a mean age of 7 years. The CT acquisition, contrast, and reconstruction protocols varied across the scanner models and patients, with specifications available in the DICOM headers. Expert contours were manually labeled for up to 29 organ structures per subject. Not all contours are available for all subjects, due to limited field of view or unreliable contouring due to high noise. Data format and usage notes The data are available on The Cancer Imaging Archive (TCIA_ () under the collection Pediatric-CT-SEG. The axial CT image slices for each subject are available in DICOM format. The expert contours are stored in a single DICOM RTSTRUCT file for each subject. The contour names are listed in Table 2. Potential applications This dataset will enable the evaluation and development of organ autosegmentation algorithms for pediatric populations, which exhibit variations in organ shape and size across age. Automated organ segmentation from CT images has numerous applications including radiation therapy, diagnostic tasks, surgical planning, and patient-specific organ dose estimation.

Fenfluramine (2.5 and 5 mg/kg) significantly suppressed the food intake of rats following food deprivation, administration of 2-deoxy-D-glucose (2DG), and during tail pressure. This suggests that fenfluramine has relatively general anorectic potency. Other “serotonergic” anorectics were studied for comparison. In a second experiment we determined that norfenfluramine and quipazine greatly suppressed food intake following food deprivation but, at the same doses, had relatively small effects on water intake following water deprivation. This was true for intraperitoneal and cerebroventricular routes of administration. The data have relevance for specificity of action of these agents and for the possible contribution of dopamine antagonist properties.

Dual-energy CT (DECT) with scans over limited-angular ranges (LARs) may allow reductions in scan time and radiation dose and avoidance of possible collision between the moving parts of a scanner and the imaged object. The beam-hardening (BH) and LAR effects are two sources of image artifacts in DECT with LAR data. In this work, we investigate a two-step method to correct for both BH and LAR artifacts in order to yield accurate image reconstruction in DECT with LAR data. From low- and high-kVp LAR data in DECT, we first use a data-domain decomposition (DDD) algorithm to obtain LAR basis data with the non-linear BH effect corrected for. We then develop and tailor a directional-total-variation (DTV) algorithm to reconstruct from the LAR basis data obtained basis images with the LAR effect compensated for. Finally, using the basis images reconstructed, we create virtual monochromatic images (VMIs), and estimate physical quantities such as iodine concentrations and effective atomic numbers within the object imaged. We conduct numerical studies using two digital phantoms of different complexity levels and types of structures. LAR data of low- and high-kVp are generated from the phantoms over both single-arc (SA) and two-orthogonal-arc (TOA) LARs ranging from 14∘ to 180∘. Visual inspection and quantitative assessment of VMIs obtained reveal that the two-step method proposed can yield VMIs in which both BH and LAR artifacts are reduced, and estimation accuracy of physical quantities is improved. In addition, concerning SA and TOA scans with the same total LAR, the latter is shown to yield more accurate images and physical quantity estimations than the former. We investigate a two-step method that combines the DDD and DTV algorithms to correct for both BH and LAR artifacts in image reconstruction, yielding accurate VMIs and estimations of physical quantities, from low- and high-kVp LAR data in DECT. The results and knowledge acquired in the work on accurate image reconstruction in LAR DECT may give rise to further understanding and insights into the practical design of LAR scan configurations and reconstruction procedures for DECT applications.

Background Calibration of photon-counting detectors (PCDs) is necessary for quantitatively accurate spectral computed tomography (CT), but the calibration process can be complicated by nonlinear flux-dependent physical factors such as pulse pile-up. Purpose This work develops a method for spectral sensitivity calibration of a PCD-based spectral CT system that incorporates nonlinear flux dependence and can thus be employed at high photon flux. Methods A calibration model for the spectral response and polynomial flux dependence is proposed, which incorporates prior x-ray source spectrum and PCD models and that has a small set of parameters for adjusting to the spectral CT system of interest. The model parameters are determined by fitting transmission data from a known object of known composition: a step-wedge phantom composed of different thicknesses of aluminum, a bone equivalent, and polymethyl methacrylate (PMMA), a soft-tissue equivalent. This fitting employs Tikhonov regularization, and the regularization strength and the polynomial order for the intensity modeling are determined by bias and variance analysis. The spectral calibration and nonlinear intensity correction is validated on transmission measurements through a third material, Teflon, at different x-ray photon flux levels. Results The nonlinear intensity dependence is determined to be accurately accounted for with a third-order polynomial. The calibrated spectral CT model accurately predicts Teflon transmission to within 1% for flux levels up to 50% of the detector maximum. Conclusions The proposed PCD calibration method enables accurate physical modeling necessary for quantitative imaging in spectral CT. Furthermore, the model applies to high flux settings so that acquisition times will not be limited by restricting the spectral CT system to low flux levels.

Dual-kV is a spectral CT modality that is currently available clinically, which traditionally employs energy integrating detectors, and in some cases acquire spectral data that is unregistered. A spectral method developed for photon-counting detectors, the constrained one-step spectral CT image reconstruction method (cOSSCIR), is designed to perform material decomposition directly from a set of spectral measurements. As such, the spectral sinograms need not be registered. Within the framework of cOSSCIR, dual-kV data is just two-window spectral CT. Furthermore, reconstructing unregistered data, such as that acquired by fast kV switching and slow kV switching systems, provides a potential extension of sOSSCIR to reconstructing clinical data. In this investigation we demonstrate cOSSCIR on unregistered dual-kV protocols using simulations and experimental phantom studies. Our results suggest that cOSSCIR can accurately recover the basis material maps using slow- and rapid-kV data compared to fully registered reconstructions.

Purpose This study developed and evaluated a fully convolutional network (FCN) for pediatric CT organ segmentation and investigated the generalizability of the FCN across image heterogeneities such as CT scanner model protocols and patient age. We also evaluated the autosegmentation models as part of a software tool for patient-specific CT dose estimation. Methods A collection of 359 pediatric CT datasets with expert organ contours were used for model development and evaluation. Autosegmentation models were trained for each organ using a modified FCN 3D V-Net. An independent test set of 60 patients was withheld for testing. To evaluate the impact of CT scanner model protocol and patient age heterogeneities, separate models were trained using a subset of scanner model protocols and pediatric age groups. Train and test sets were split to answer questions about the generalizability of pediatric FCN autosegmentation models to unseen age groups and scanner model protocols, as well as the merit of scanner model protocol or age-group-specific models. Finally, the organ contours resulting from the autosegmentation models were applied to patient-specific dose maps to evaluate the impact of segmentation errors on organ dose estimation. Results Results demonstrate that the autosegmentation models generalize to CT scanner acquisition and reconstruction methods which were not present in the training dataset. While models are not equally generalizable across age groups, age-group-specific models do not hold any advantage over combining heterogeneous age groups into a single training set. Dice similarity coefficient (DSC) and mean surface distance results are presented for 19 organ structures, for example, median DSC of 0.52 (duodenum), 0.74 (pancreas), 0.92 (stomach), and 0.96 (heart). The FCN models achieve a mean dose error within 5% of expert segmentations for all 19 organs except for the spinal canal, where the mean error was 6.31%. Conclusions Overall, these results are promising for the adoption of FCN autosegmentation models for pediatric CT, including applications for patient-specific CT dose estimation.

Objective This work aimed to retrospectively evaluate the potential of dose reduction on chest computed tomography (CT) examinations by reducing the longitudinal scan length for patients positive for coronavirus disease 2019 (COVID-19). Methods This study used the Personalized Rapid Estimation of Dose in CT (PREDICT) tool to estimate patient-specific organ doses from CT image data. The PREDICT is a research tool that combines a linear Boltzmann transport equation solver for radiation dose map generation with deep learning algorithms for organ contouring. Computed tomography images from 74 subjects in the Medical Imaging Data Resource Center–RSNA International COVID-19 Open Radiology Database data set (chest CT of adult patients positive for COVID-19), which included expert annotations including “infectious opacities,” were analyzed. First, the full z-scan length of the CT image data set was evaluated. Next, the z-scan length was reduced from the left hemidiaphragm to the top of the aortic arch. Generic dose reduction based on dose length product (DLP) and patient-specific organ dose reductions were calculated. The percentage of infectious opacities excluded from the reduced z-scan length was used to quantify the effect on diagnostic utility. Results Generic dose reduction, based on DLP, was 69%. The organ dose reduction ranged from approximately equal to 18% (breasts) to approximately equal to 64% (bone surface and bone marrow). On average, 12.4% of the infectious opacities were not included in the reduced z-coverage, per patient, of which 5.1% were above the top of the arch and 7.5% below the left hemidiaphragm. Conclusions Limiting z-scan length of chest CTs reduced radiation dose without significantly compromising diagnostic utility in COVID-19 patients. The PREDICT demonstrated that patient-specific organ dose reductions varied from generic dose reduction based on DLP.

Microscopy with ultraviolet surface excitation (MUSE) is increasingly studied for intraoperative assessment of tumor margins during breast-conserving surgery to reduce the re-excision rate. Here we report a two-step classification approach using texture analysis of MUSE images to automate the margin detection. A study dataset consisting of MUSE images from 66 human breast tissues was constructed for model training and validation. Features extracted using six texture analysis methods were investigated for tissue characterization, and a support vector machine was trained for binary classification of image patches within a full image based on selected feature subsets. A weighted majority voting strategy classified a sample as tumor or normal. Using the eight most predictive features ranked by the maximum relevance minimum redundancy and Laplacian scores methods has achieved a sample classification accuracy of 92.4% and 93.0%, respectively. Local binary pattern alone has achieved an accuracy of 90.3%.