Improved efficacy of pembrolizumab combined with soluble EphB4-albumin in HPV-negative EphrinB2 positive head neck squamous cell carcinoma.

OBJECTIVE: Patients with relapsed or metastatic head and neck squamous cell carcinoma (HNSCC) after primary local therapy have low response rates with cetuximab, systemic chemotherapy or check point inhibitor therapy. Novel combination therapies with the potential to improve outcomes for patients with HNSCC is an area of high unmet need. METHODS: This is a phase II single-arm clinical trial of locally advanced or metastatic HNSCC patients treated with a combination of soluble EphB4-human serum albumin (sEphB4-HSA) fusion protein and pembrolizumab after platinum-based chemotherapy with up to 2 prior lines of treatment. The primary endpoints were safety and tolerability and the primary efficacy endpoint was overall response rate (ORR). Secondary endpoints included progression free survival (PFS) and overall survival (OS). HPV status and EphrinB2 expression were evaluated for outcome. RESULTS: Twenty-five patients were enrolled. Median follow up was 40.4 months (range 9.8 - 40.4). There were 6 responders (ORR 24%). There were 5 responders in the 11 HPV-negative and EphrinB2 positive patients, (ORR 45%) with 2 of these patients achieving a complete response (CR). The median PFS in HPV-negative/EphrinB2 positive patients was 3.2 months (95% CI 1.1, 7.3). Median OS in HPV-negative/EphrinB2 positive patients was 10.9 months (95% CI 2.0, 13.7). Hypertension, transaminitis and fatigue were the most common toxicities. DISCUSSION: The combination of sEphB4-HSA and pembrolizumab has a favorable toxicity profile and favorable activity particularly among HPV-negative EphrinB2 positive patients with HNSCC.

PCa-RadHop: A transparent and lightweight feed-forward method for clinically significant prostate cancer segmentation.

Prostate Cancer is one of the most frequently occurring cancers in men, with a low survival rate if not early diagnosed. PI-RADS reading has a high false positive rate, thus increasing the diagnostic incurred costs and patient discomfort. Deep learning (DL) models achieve a high segmentation performance, although require a large model size and complexity. Also, DL models lack of feature interpretability and are perceived as "black-boxes" in the medical field. PCa-RadHop pipeline is proposed in this work, aiming to provide a more transparent feature extraction process using a linear model. It adopts the recently introduced Green Learning (GL) paradigm, which offers a small model size and low complexity. PCa-RadHop consists of two stages: Stage-1 extracts data-driven radiomics features from the bi-parametric Magnetic Resonance Imaging (bp-MRI) input and predicts an initial heatmap. To reduce the false positive rate, a subsequent stage-2 is introduced to refine the predictions by including more contextual information and radiomics features from each already detected Region of Interest (ROI). Experiments on the largest publicly available dataset, PI-CAI, show a competitive performance standing of the proposed method among other deep DL models, achieving an area under the curve (AUC) of 0.807 among a cohort of 1,000 patients. Moreover, PCa-RadHop maintains orders of magnitude smaller model size and complexity.

Deep Learning De-Noising Improves CT Perfusion Image Quality in the Setting of Lower Contrast Dosing: A Feasibility Study.

BACKGROUND AND PURPOSE: Considering recent iodinated contrast media (ICM) shortages, this study compared reduced ICM and standard dose CTP acquisitions, and the impact of deep learning (DL)-denoising on CTP image quality in preclinical and clinical studies. MATERIALS AND METHODS: Twelve swine underwent 9 CTP exams each, performed at combinations of 3 different X-ray (37, 67, and 127mAs) and ICM doses (10, 15, and 20mL). Clinical CTP acquisitions performed before and during the ICM shortage and protocol change (from 40 mL to 30 mL) were retrospectively included. Eleven patients with reduced ICM dose and 11 propensity-score-matched controls with standard ICM dose were included. A Residual Encoder-Decoder Convolutional-Neural-Network (RED-CNN) was trained for CTP denoising using K-space-Weighted Image Average (KWIA) filtered CTP images as the target. The standard, RED-CNN denoised, and KWIA noise-filtered images for animal and human studies were compared for quantitative SNR and qualitative image evaluation. RESULTS: The SNR of animal CTP images decreased with reductions in ICM and mAs doses. Contrast dose reduction had a greater effect on SNR than mAs reduction. Noise-filtering by KWIA and RED-CNN denoising progressively improved SNR of CTP maps, with RED-CNN resulting in the highest SNR. The SNR of clinical CTP images was generally lower with reduced ICM dose, which was improved by KWIA and RED-CNN denoising (p<0.05). Qualitative readings consistently rated RED-CNN denoised CTP as best quality, followed by KWIA and then standard CTP images. CONCLUSIONS: DL-denoising can improve image quality for low ICM CTP protocols, and could approximate standard ICM dose CTP, in addition to potentially improving image quality for low mAs acquisitions. ABBREVIATIONS: ICM=iodinated contrast media; DL=deep learning; KWIA=k-space weighted image average; LCD=low-contrast dose; SCD=standard contrast dose; RED-CNN=Residual Encoder-Decoder Convolutional Neural Network; PSNR=Peak Signal to Noise Ratio; RMSE=Root Mean Squared Error; SSIM=Structural Similarity Index.

A nomogram to predict the absence of clinically significant prostate cancer in males with negative MRI

ABSTRACT Purpose: To create a nomogram to predict the absence of clinically significant prostate cancer (CSPCa) in males with non-suspicion multiparametric magnetic resonance imaging (mpMRI) undergoing prostate biopsy (PBx). Materials and Methods: We identified consecutive patients who underwent 3T mpMRI followed by PBx for suspicion of PCa or surveillance follow-up. All patients had Prostate Imaging Reporting and Data System score 1-2 (negative mpMRI). CSPCa was defined as Grade Group ≥2. Multivariate logistic regression analysis was performed via backward elimination. Discrimination was evaluated with area under the receiver operating characteristic (AUROC). Internal validation with 1,000x bootstrapping for estimating the optimism corrected AUROC. Results: Total 327 patients met inclusion criteria. The median (IQR) age and PSA density (PSAD) were 64 years (58-70) and 0.10 ng/mL2 (0.07-0.15), respectively. Biopsy history was as follows: 117 (36%) males were PBx-naive, 130 (40%) had previous negative PBx and 80 (24%) had previous positive PBx. The majority were White (65%); 6% of males self-reported Black. Overall, 44 (13%) patients were diagnosed with CSPCa on PBx. Black race, history of previous negative PBx and PSAD ≥0.15ng/mL2 were independent predictors for CSPCa on PBx and were included in the nomogram. The AUROC of the nomogram was 0.78 and the optimism corrected AUROC was 0.75. Conclusions: Our nomogram facilitates evaluating individual probability of CSPCa on PBx in males with PIRADS 1-2 mpMRI and may be used to identify those in whom PBx may be safely avoided. Black males have increased risk of CSPCa on PBx, even in the setting of PIRADS 1-2 mpMRI

Prophylactic Use of Biologic Mesh in Ileal Conduit (PUBMIC): A Randomized Clinical Trial.

PURPOSE: We assessed the effect of prophylactic biologic mesh on parastomal hernia (PSH) development in patients undergoing cystectomy and ileal conduit (IC). MATERIALS AND METHODS: This phase 3, randomized, controlled trial (NCT02439060) included 146 patients who underwent cystectomy and IC at the University of Southern California between 2015 and 2021. Follow-ups were physical exam and CT every 4 to 6 months up to 2 years. Patients were randomized 1:1 to receive FlexHD prophylactic biological mesh using sublay intraperitoneal technique vs standard IC. The primary end point was time to radiological PSH, and secondary outcomes included clinical PSH with/without surgical intervention and mesh-related complications. RESULTS: The 2 arms were similar in terms of baseline clinical features. All surgeries and mesh placements were performed without any intraoperative complications. Median operative time was 31 minutes longer in patients who received mesh, yet with no statistically significant difference (363 vs 332 minutes, P = .16). With a median follow-up of 24 months, radiological and clinical PSHs were detected in 37 (18 mesh recipients vs 19 controls) and 16 (8 subjects in both arms) patients, with a median time to radiological and clinical PSH of 8.3 and 15.5 months, respectively. No definite mesh-related adverse events were reported. Five patients (3 in the mesh and 2 in the control arm) required surgical PSH repair. Radiological PSH-free survival rates in the mesh and control groups were 74% vs 75% at 1 year and 69% vs 62% at 2 years. CONCLUSIONS: Implementation of biologic mesh at the time of IC construction is safe without significant protective effects within 2 years following surgery.

Technical and clinical considerations of a physical liver phantom for CT radiomics analysis.

OBJECTIVE: This study identifies key characteristics to help build a physical liver computed tomography (CT) phantom for radiomics harmonization; particularly, the higher-order texture metrics. MATERIALS AND METHODS: CT scans of a radiomics phantom comprising of 18 novel 3D printed inserts with varying size, shape, and material combinations were acquired on a 64-slice CT scanner (Brilliance 64, Philips Healthcare). The images were acquired at 120 kV, 250 mAs, CTDIvol of 16.36 mGy, 2 mm slice thickness, and iterative noise-reduction reconstruction (iDose, Philips Healthcare, Andover, MA). Radiomics analysis was performed using the Cancer Imaging Phenomics Toolkit (CaPTk), following automated segmentation of 3D regions of interest (ROI) of the 18 inserts. The findings were compared to three additional ROI obtained of an anthropomorphic liver phantom, a patient liver CT scan, and a water phantom, at comparable imaging settings. Percentage difference in radiomic metrics values between phantom and tissue was used to assess the biological equivalency and <10% was used to claim equivalent. RESULTS: The HU for all 18 ROI from the phantom ranged from -30 to 120 which is within clinically observed HU range of the liver, showing that our phantom material (T3-6B) is representative of biological CT tissue densities (liver) with >50% radiomic features having <10% difference from liver tissue. Based on the assessment of the Neighborhood Gray Tone Difference Matrix (NGTDM) metrics it is evident that the water phantom ROI show extreme values compared to the ROIs from the phantom. This result may further reinforce the difference between a structureless quantity such as water HU values and tissue HU values found in liver. CONCLUSION: The 3-D printed patterns of the constructed radiomics phantom cover a wide span of liver tissue textures seen in CT images. Using our results, texture metrics can be selectively harmonized to establish clinically relevant and reliable radiomics panels.

Conditional generative learning for medical image imputation.

Image imputation refers to the task of generating a type of medical image given images of another type. This task becomes challenging when the difference between the available images, and the image to be imputed is large. In this manuscript, one such application is considered. It is derived from the dynamic contrast enhanced computed tomography (CECT) imaging of the kidneys: given an incomplete sequence of three CECT images, we are required to impute the missing image. This task is posed as one of probabilistic inference and a generative algorithm to generate samples of the imputed image, conditioned on the available images, is developed, trained, and tested. The output of this algorithm is the "best guess" of the imputed image, and a pixel-wise image of variance in the imputation. It is demonstrated that this best guess is more accurate than those generated by other, deterministic deep-learning based algorithms, including ones which utilize additional information and more complex loss terms. It is also shown that the pixel-wise variance image, which quantifies the confidence in the reconstruction, can be used to determine whether the result of the imputation meets a specified accuracy threshold and is therefore appropriate for a downstream task.

Pseudocirrhosis: A Case Series with Clinical and Radiographic Correlation and Review of the Literature.

BACKGROUND AND AIMS: Pseudocirrhosis is a poorly understood acquired morphologic change of the liver that occurs in the setting of metastatic malignancy and radiographically resembles cirrhosis. Pseudocirrhosis has been primarily described in metastatic breast carcinoma, with few case reports arising from other primary malignancies. We present 29 cases of pseudocirrhosis, including several cases from primary malignancies not previously described. METHODS: Radiologic, clinical, demographic, and biomedical data were collected retrospectively and analyzed. We compared clinical and radiologic characteristics and outcomes between patients with pseudocirrhosis arising in metastatic breast cancer and non-breast primary malignancies. RESULTS: Among the 29 patients, 14 had breast cancer and 15 had non-breast primaries including previously never reported primaries associated with pseudocirrhosis, melanoma, renal cell carcinoma, appendiceal carcinoid, and cholangiocarcinoma. Median time from cancer diagnosis to development of pseudocirrhosis was 80.8 months for patients with primary breast cancer and 29.8 months for non-breast primary (p = 0.02). Among all patients, 15 (52%) had radiographic features of portal hypertension. Radiographic evidence of portal hypertension was identified in 28.6% of breast cancer patients, compared to 73.3% of those with non-breast malignancies (p = 0.03). CONCLUSION: Pseudocirrhosis has most commonly been described in the setting of metastatic breast cancer but occurs in any metastatic disease to the liver. Our study suggests that portal hypertensive complications are more common in the setting of non-breast primary cancers than in metastatic breast cancer. Prior exposure to multiple chemotherapeutic agents, and agents known to cause sinusoidal injury, is a common feature but not essential for the development of pseudocirrhosis.

A nomogram to predict the absence of clinically significant prostate cancer in males with negative MRI

PURPOSE: To create a nomogram to predict the absence of clinically significant prostate cancer (CSPCa) in males with non-suspicion multiparametric magnetic resonance imaging (mpMRI) undergoing prostate biopsy (PBx). MATERIALS AND METHODS: We identified consecutive patients who underwent 3T mpMRI followed by PBx for suspicion of PCa or surveillance follow-up. All patients had Prostate Imaging Reporting and Data System score 1-2 (negative mpMRI). CSPCa was defined as Grade Group ≥2. Multivariate logistic regression analysis was performed via backward elimination. Discrimination was evaluated with area under the receiver operating characteristic (AUROC). Internal validation with 1,000x bootstrapping for estimating the optimism corrected AUROC. RESULTS: Total 327 patients met inclusion criteria. The median (IQR) age and PSA density (PSAD) were 64 years (58-70) and 0.10 ng/mL2 (0.07-0.15), respectively. Biopsy history was as follows: 117 (36%) males were PBx-naive, 130 (40%) had previous negative PBx and 80 (24%) had previous positive PBx. The majority were White (65%); 6% of males self-reported Black. Overall, 44 (13%) patients were diagnosed with CSPCa on PBx. Black race, history of previous negative PBx and PSAD ≥0.15ng/mL2 were independent predictors for CSPCa on PBx and were included in the nomogram. The AUROC of the nomogram was 0.78 and the optimism corrected AUROC was 0.75. CONCLUSIONS: Our nomogram facilitates evaluating individual probability of CSPCa on PBx in males with PIRADS 1-2 mpMRI and may be used to identify those in whom PBx may be safely avoided. Black males have increased risk of CSPCa on PBx, even in the setting of PIRADS 1-2 mpMRI.

Radiomics Correlation to CD68+ Tumor-Associated Macrophages in Clear Cell Renal Cell Carcinoma.

INTRODUCTION: Renal cell carcinoma (RCC) is the ninth most common cancer worldwide, with clear cell RCC (ccRCC) being the most frequent histological subtype. The tumor immune microenvironment (TIME) of ccRCC is an important factor to guide treatment, but current assessments are tissue-based, which can be time-consuming and resource-intensive. In this study, we used radiomics extracted from clinically performed computed tomography (CT) as a noninvasive surrogate for CD68 tumor-associated macrophages (TAMs), a significant component of ccRCC TIME. METHODS: TAM population was measured by CD68+/PanCK+ ratio and tumor-TAM clustering was measured by normalized K function calculated from multiplex immunofluorescence (mIF). A total of 1,076 regions on mIF slides from 78 patients were included. Radiomic features were extracted from multiphase CT of the ccRCC tumor. Statistical machine learning models, including random forest, Adaptive Boosting, and ElasticNet, were used to predict TAM population and tumor-TAM clustering. RESULTS: The best models achieved an area under the ROC curve of 0.81 (95% CI: [0.69, 0.92]) for TAM population and 0.77 (95% CI: [0.66, 0.88]) for tumor-TAM clustering, respectively. CONCLUSION: Our study demonstrates the potential of using CT radiomics-derived imaging markers as a surrogate for assessment of TAM in ccRCC for real-time treatment response monitoring and patient selection for targeted therapies and immunotherapies.

Investigating the role of imaging factors in the variability of CT-based texture analysis metrics.

OBJECTIVE: This study assesses the robustness of first-order radiomic texture features namely interquartile range (IQR), coefficient of variation (CV) and standard deviation (SD) derived from computed tomography (CT) images by varying dose, reconstruction algorithms and slice thickness using scans of a uniform water phantom, a commercial anthropomorphic liver phantom, and a human liver in-vivo. MATERIALS AND METHODS: Scans were acquired on a 16 cm detector GE Revolution Apex Edition CT scanner with variations across three different nominal slice thicknesses: 0.625, 1.25, and 2.5 mm, three different dose levels: CTDIvol of 13.86 mGy for the standard dose, 40% reduced dose and 60% reduced dose and two different reconstruction algorithms: a deep learning image reconstruction (DLIR-high) algorithm and a hybrid iterative reconstruction (IR) algorithm ASiR-V50% (AV50) were explored, varying one at a time. To assess the effect of non-linear modifications of images by AV50 and DLIR-high, images of the water phantom were also reconstructed using filtered back projection (FBP). Quantitative measures of IQR, CV and SD were extracted from twelve pre-selected, circular (1 cm diameter) regions of interest (ROIs) capturing different texture patterns across all scans. RESULTS: Across all scans, imaging, and reconstruction settings, CV, IQR and SD were observed to increase with reduction in dose and slice thickness. An exception to this observation was found when using FBP reconstruction. Lower values of CV, IQR and SD were observed in DLIR-high reconstructions compared to AV50 and FBP. The Poisson statistics were more stringently noted in FBP than DLIR-high and AV50, due to the non-linear nature of the latter two algorithms. CONCLUSION: Variation in image noise due to dose reduction algorithms, tube current, and slice thickness show a consistent trend across phantom and patient scans. Prospective evaluation across multiple centers, scanners and imaging protocols is needed for establishing quality assurance standards of radiomics.

Diversity in Renal Mass Data Cohorts: Implications for Urology AI Researchers.

INTRODUCTION: We examine the heterogeneity and distribution of the cohort populations in two publicly used radiological image cohorts, the Cancer Genome Atlas Kidney Renal Clear Cell Carcinoma (TCIA TCGA KIRC) collection and 2019 MICCAI Kidney Tumor Segmentation Challenge (KiTS19), and deviations in real-world population renal cancer data from the National Cancer Database (NCDB) Participant User Data File (PUF) and tertiary center data. PUF data are used as an anchor for prevalence rate bias assessment. Specific gene expression and, therefore, biology of RCC differ by self-reported race, especially between the African American and Caucasian populations. AI algorithms learn from datasets, but if the dataset misrepresents the population, reinforcing bias may occur. Ignoring these demographic features may lead to inaccurate downstream effects, thereby limiting the translation of these analyses to clinical practice. Consciousness of model training biases is vital to patient care decisions when using models in clinical settings. METHODS: Data elements evaluated included gender, demographics, reported pathologic grading, and cancer staging. American Urological Association risk levels were used. Poisson regression was performed to estimate the population-based and sample-specific estimation for prevalence rate and corresponding 95% confidence interval. SAS 9.4 was used for data analysis. RESULTS: Compared to PUF, KiTS19 and TCGA KIRC oversampled Caucasian by 9.5% (95% CI, -3.7 to 22.7%) and 15.1% (95% CI, 1.5 to 28.8%), undersampled African American by -6.7% (95% CI, -10% to -3.3%), and -5.5% (95% CI, -9.3% to -1.8%). Tertiary also undersampled African American by -6.6% (95% CI, -8.7% to -4.6%). The tertiary cohort largely undersampled aggressive cancers by -14.7% (95% CI, -20.9% to -8.4%). No statistically significant difference was found among PUF, TCGA, and KiTS19 in aggressive rate; however, heterogeneities in risk are notable. CONCLUSION: Heterogeneities between cohorts need to be considered in future AI training and cross-validation for renal masses.

Feminizing Adrenocortical Tumor with Multiple Recurrences: A Case Report.

Feminizing adrenocortical tumors (FATs) are exceptionally rare primary adrenal neoplasms that cause high estrogen and low testosterone levels. They are most common in adult males, typically presenting with gynecomastia, hypogonadism, and weight loss. They are almost always malignant, with a poor prognosis and a high recurrence rate. We report a case of a 35-year-old man with an adrenal FAT with high estrogen (181 pg/mL) and low testosterone (37 ng/dL) who presented with gynecomastia, erectile dysfunction, subclinical Cushing syndrome, and pain localizing to different regions of the torso. There was no evidence of metastatic disease initially as seen by visualization of a well-marginated mass on computed tomography scan. Surgical resection of the FAT was performed, and the mass was confirmed to be a low-grade tumor. Clinical symptoms were resolved after surgery. Despite complete resection with negative margins, the patient subsequently had two separate local metastatic recurrences within a few years, treated with a combination of further surgery and medical intervention. This case highlights the unique features of an exceedingly rare adrenal tumor and stresses the importance of early detection and vigilant surveillance following resection due to high recurrence rates.

Localized Multifocal Retroperitoneal Ganglioneuroma with an Infiltrative Appearance on Imaging: A Case Report.

Multifocal ganglioneuromas are characterized by the presence of multiple benign neuroepithelial tumor nodules and are less common than solitary tumors. A small percentage of ganglioneuromas present with a fatty appearance. Only a few cases of multifocal ganglioneuromas have been reported, due to both their rarity and minimal symptomatic presentation; therefore, generalizations about risk factors and predictive markers are very difficult. Here, we report a case of multifocal retroperitoneal ganglioneuroma with an infiltrative appearance on computed tomography (CT). The tumor demonstrated slow growth on multiple imaging studies and was associated with abdominal and flank pain. The aggressive appearance eventually led to surgical resection 18 months after the initial incidental finding on CT. Postsurgical analysis of the tumor on imaging was crucial in revealing its nodularity and infiltration, as well as for clarifying its retroperitoneal location inseparable from the adrenal gland. Histology demonstrated Schwann cells and ganglion cells without atypia or increased cellularity, and with no mitosis or necrosis seen. Our case highlights the consideration of ganglioneuroma with fatty infiltration in the differential diagnosis of a fatty tumor in the mediastinum or retroperitoneum. Additionally, our report differentiates multifocal ganglioneuroma with fatty infiltration from lipomatous ganglioneuroma on radiology and histopathology.

Comprehensive Systematic Review of Biomarkers in Metastatic Renal Cell Carcinoma: Predictors, Prognostics, and Therapeutic Monitoring.

BACKGROUND: Challenges remain in determining the most effective treatment strategies and identifying patients who would benefit from adjuvant or neoadjuvant therapy in renal cell carcinoma. The objective of this review is to provide a comprehensive overview of biomarkers in metastatic renal cell carcinoma (mRCC) and their utility in prediction of treatment response, prognosis, and therapeutic monitoring in patients receiving systemic therapy for metastatic disease. METHODS: A systematic literature search was conducted using the PubMed database for relevant studies published between January 2017 and December 2022. The search focused on biomarkers associated with mRCC and their relationship to immune checkpoint inhibitors, targeted therapy, and VEGF inhibitors in the adjuvant, neoadjuvant, and metastatic settings. RESULTS: The review identified various biomarkers with predictive, prognostic, and therapeutic monitoring potential in mRCC. The review also discussed the challenges associated with anti-angiogenic and immune-checkpoint monotherapy trials and highlighted the need for personalized therapy based on molecular signatures. CONCLUSION: This comprehensive review provides valuable insights into the landscape of biomarkers in mRCC and their potential applications in prediction of treatment response, prognosis, and therapeutic monitoring. The findings underscore the importance of incorporating biomarker assessment into clinical practice to guide treatment decisions and improve patient outcomes in mRCC.

Imaging in Upper Tract Urothelial Carcinoma: A Review.

Medical imaging is a critical tool in the detection, staging, and treatment planning of upper urinary tract urothelial carcinoma (UTUC). This article reviews the strengths and weaknesses of the different imaging techniques and modalities available clinically. This includes multidetector computed tomography (CT), multiparametric magnetic resonance imaging (MRI), ultrasound (US), and positron emission tomography (PET) for the detection, staging, and management of UTUC. In addition, we review the imaging techniques that are being developed and are on the horizon but have not yet made it to clinical practice. Firstly, we review the imaging findings of primary UTUC and the techniques across multiple modalities. We then discuss imaging findings of metastatic disease. Lastly, we describe the role of imaging in the surveillance after resection of primary UTUC based upon current guidelines.

Spatial assessments in texture analysis: what the radiologist needs to know.

To date, studies investigating radiomics-based predictive models have tended to err on the side of data-driven or exploratory analysis of many thousands of extracted features. In particular, spatial assessments of texture have proven to be especially adept at assessing for features of intratumoral heterogeneity in oncologic imaging, which likewise may correspond with tumor biology and behavior. These spatial assessments can be generally classified as spatial filters, which detect areas of rapid change within the grayscale in order to enhance edges and/or textures within an image, or neighborhood-based methods, which quantify gray-level differences of neighboring pixels/voxels within a set distance. Given the high dimensionality of radiomics datasets, data dimensionality reduction methods have been proposed in an attempt to optimize model performance in machine learning studies; however, it should be noted that these approaches should only be applied to training data in order to avoid information leakage and model overfitting. While area under the curve of the receiver operating characteristic is perhaps the most commonly reported assessment of model performance, it is prone to overestimation when output classifications are unbalanced. In such cases, confusion matrices may be additionally reported, whereby diagnostic cut points for model predicted probability may hold more clinical significance to clinical colleagues with respect to related forms of diagnostic testing.

Deep learning image segmentation approaches for malignant bone lesions: a systematic review and meta-analysis.

Introduction: Image segmentation is an important process for quantifying characteristics of malignant bone lesions, but this task is challenging and laborious for radiologists. Deep learning has shown promise in automating image segmentation in radiology, including for malignant bone lesions. The purpose of this review is to investigate deep learning-based image segmentation methods for malignant bone lesions on Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and Positron-Emission Tomography/CT (PET/CT). Method: The literature search of deep learning-based image segmentation of malignant bony lesions on CT and MRI was conducted in PubMed, Embase, Web of Science, and Scopus electronic databases following the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). A total of 41 original articles published between February 2017 and March 2023 were included in the review. Results: The majority of papers studied MRI, followed by CT, PET/CT, and PET/MRI. There was relatively even distribution of papers studying primary vs. secondary malignancies, as well as utilizing 3-dimensional vs. 2-dimensional data. Many papers utilize custom built models as a modification or variation of U-Net. The most common metric for evaluation was the dice similarity coefficient (DSC). Most models achieved a DSC above 0.6, with medians for all imaging modalities between 0.85-0.9. Discussion: Deep learning methods show promising ability to segment malignant osseous lesions on CT, MRI, and PET/CT. Some strategies which are commonly applied to help improve performance include data augmentation, utilization of large public datasets, preprocessing including denoising and cropping, and U-Net architecture modification. Future directions include overcoming dataset and annotation homogeneity and generalizing for clinical applicability.