ABSTRACT
PURPOSE: To determine if pelvic/ovarian and omental lesions of ovarian cancer can be reliably segmented on computed tomography (CT) using fully automated deep learning-based methods. METHODS: A deep learning model for the two most common disease sites of high-grade serous ovarian cancer lesions (pelvis/ovaries and omentum) was developed and compared against the well-established "no-new-Net" framework and unrevised trainee radiologist segmentations. A total of 451 CT scans collected from four different institutions were used for training (n = 276), evaluation (n = 104) and testing (n = 71) of the methods. The performance was evaluated using the Dice similarity coefficient (DSC) and compared using a Wilcoxon test. RESULTS: Our model outperformed no-new-Net for the pelvic/ovarian lesions in cross-validation, on the evaluation and test set by a significant margin (p values being 4 × 10-7, 3 × 10-4, 4 × 10-2, respectively), and for the omental lesions on the evaluation set (p = 1 × 10-3). Our model did not perform significantly differently in segmenting pelvic/ovarian lesions (p = 0.371) compared to a trainee radiologist. On an independent test set, the model achieved a DSC performance of 71 ± 20 (mean ± standard deviation) for pelvic/ovarian and 61 ± 24 for omental lesions. CONCLUSION: Automated ovarian cancer segmentation on CT scans using deep neural networks is feasible and achieves performance close to a trainee-level radiologist for pelvic/ovarian lesions. RELEVANCE STATEMENT: Automated segmentation of ovarian cancer may be used by clinicians for CT-based volumetric assessments and researchers for building complex analysis pipelines. KEY POINTS: ⢠The first automated approach for pelvic/ovarian and omental ovarian cancer lesion segmentation on CT images has been presented. ⢠Automated segmentation of ovarian cancer lesions can be comparable with manual segmentation of trainee radiologists. ⢠Careful hyperparameter tuning can provide models significantly outperforming strong state-of-the-art baselines.
Subject(s)
Deep Learning , Ovarian Cysts , Ovarian Neoplasms , Humans , Female , Ovarian Neoplasms/diagnostic imaging , Neural Networks, Computer , Tomography, X-Ray ComputedABSTRACT
High grade serous ovarian carcinoma (HGSOC) is a highly heterogeneous disease that typically presents at an advanced, metastatic state. The multi-scale complexity of HGSOC is a major obstacle to predicting response to neoadjuvant chemotherapy (NACT) and understanding critical determinants of response. Here we present a framework to predict the response of HGSOC patients to NACT integrating baseline clinical, blood-based, and radiomic biomarkers extracted from all primary and metastatic lesions. We use an ensemble machine learning model trained to predict the change in total disease volume using data obtained at diagnosis (n = 72). The model is validated in an internal hold-out cohort (n = 20) and an independent external patient cohort (n = 42). In the external cohort the integrated radiomics model reduces the prediction error by 8% with respect to the clinical model, achieving an AUC of 0.78 for RECIST 1.1 classification compared to 0.47 for the clinical model. Our results emphasize the value of including radiomics data in integrative models of treatment response and provide methods for developing new biomarker-based clinical trials of NACT in HGSOC.
Subject(s)
Ovarian Neoplasms , Humans , Female , Ovarian Neoplasms/diagnostic imaging , Ovarian Neoplasms/drug therapy , Ovarian Neoplasms/genetics , Neoadjuvant Therapy/methods , Biomarkers, Tumor/geneticsABSTRACT
Background: High-Grade Serous Ovarian Carcinoma (HGSOC) is the most prevalent and lethal subtype of ovarian cancer, but has a paucity of clinically-actionable biomarkers due to high degrees of multi-level heterogeneity. Radiogenomics markers have the potential to improve prediction of patient outcome and treatment response, but require accurate multimodal spatial registration between radiological imaging and histopathological tissue samples. Previously published co-registration work has not taken into account the anatomical, biological and clinical diversity of ovarian tumours. Methods: In this work, we developed a research pathway and an automated computational pipeline to produce lesion-specific three-dimensional (3D) printed moulds based on preoperative cross-sectional CT or MRI of pelvic lesions. Moulds were designed to allow tumour slicing in the anatomical axial plane to facilitate detailed spatial correlation of imaging and tissue-derived data. Code and design adaptations were made following each pilot case through an iterative refinement process. Results: Five patients with confirmed or suspected HGSOC who underwent debulking surgery between April and December 2021 were included in this prospective study. Tumour moulds were designed and 3D-printed for seven pelvic lesions, covering a range of tumour volumes (7 to 133 cm3) and compositions (cystic and solid proportions). The pilot cases informed innovations to improve specimen and subsequent slice orientation, through the use of 3D-printed tumour replicas and incorporation of a slice orientation slit in the mould design, respectively. The overall research pathway was compatible with implementation within the clinically determined timeframe and treatment pathway for each case, involving multidisciplinary clinical professionals from Radiology, Surgery, Oncology and Histopathology Departments. Conclusions: We developed and refined a computational pipeline that can model lesion-specific 3D-printed moulds from preoperative imaging for a variety of pelvic tumours. This framework can be used to guide comprehensive multi-sampling of tumour resection specimens.
ABSTRACT
Background: Pathological response to neoadjuvant treatment for patients with high-grade serous ovarian carcinoma (HGSOC) is assessed using the chemotherapy response score (CRS) for omental tumor deposits. The main limitation of CRS is that it requires surgical sampling after initial neoadjuvant chemotherapy (NACT) treatment. Earlier and non-invasive response predictors could improve patient stratification. We developed computed tomography (CT) radiomic measures to predict neoadjuvant response before NACT using CRS as a gold standard. Methods: Omental CT-based radiomics models, yielding a simplified fully interpretable radiomic signature, were developed using Elastic Net logistic regression and compared to predictions based on omental tumor volume alone. Models were developed on a single institution cohort of neoadjuvant-treated HGSOC (n = 61; 41% complete response to NCT) and tested on an external test cohort (n = 48; 21% complete response). Results: The performance of the comprehensive radiomics models and the fully interpretable radiomics model was significantly higher than volume-based predictions of response in both the discovery and external test sets when assessed using G-mean (geometric mean of sensitivity and specificity) and NPV, indicating high generalizability and reliability in identifying non-responders when using radiomics. The performance of a fully interpretable model was similar to that of comprehensive radiomics models. Conclusions: CT-based radiomics allows for predicting response to NACT in a timely manner and without the need for abdominal surgery. Adding pre-NACT radiomics to volumetry improved model performance for predictions of response to NACT in HGSOC and was robust to external testing. A radiomic signature based on five robust predictive features provides improved clinical interpretability and may thus facilitate clinical acceptance and application.
ABSTRACT
Leiomyomas are the most common benign tumors of the uterus. On the opposite side, leiomyosarcomas are rare malignant uterine tumors that account for a significant proportion of uterine cancer deaths. Especially when large and degenerated, leiomyomas and leiomyoma variants can have overlapping imaging characteristics with those of leiomyosarcomas. Although not always possible, it is paramount to be able to differentiate between leiomyomas and leiomyosarcomas on imaging, as the therapeutic management can differ. This pictorial review aims to familiarize radiologists with imaging features of leiomyomas and various types of leiomyoma degeneration and variants, together with their pathology correlates.
Subject(s)
Leiomyoma/diagnostic imaging , Leiomyoma/pathology , Leiomyosarcoma/diagnostic imaging , Leiomyosarcoma/pathology , Magnetic Resonance Imaging/methods , Uterine Neoplasms/diagnostic imaging , Uterine Neoplasms/pathology , Diagnosis, Differential , Female , Humans , Radiology Information Systems , Uterus/diagnostic imaging , Uterus/pathologyABSTRACT
MRI was recently included as a standard pre-operative diagnostic tool for patients with endometrial cancer. MR findings allow a better risk assessment and ultimately guides the surgical planning. Therefore, it is vital that the radiological interpretation is as accurate as possible. This requires essential knowledge regarding the appropriate MRI protocol, as well as different appearances of the endometrium, ranging from normal peri- and post-menopausal changes, benign findings (e.g. endometrial hyperplasia, polyp, changes due to exogenous hormones) to common and rare endometrium-related malignancies. Furthermore, this review will emphasize the role of MRI in staging endometrial cancer patients and highlight pitfalls that could result in the underestimation or overestimation of the disease extent.
Subject(s)
Endometrial Neoplasms/diagnostic imaging , Endometrial Neoplasms/pathology , Magnetic Resonance Imaging/methods , Diagnosis, Differential , Endometrium/anatomy & histology , Endometrium/diagnostic imaging , Endometrium/pathology , Female , HumansABSTRACT
Different LI-RADS core documents were released for CEUS and for CT/MRI. Both documents rely on major and ancillary diagnostic criteria. The present paper offers an exhaustive comparison of the two documents focusing on the similarities, but especially on the differences, complementarity, and added value of imaging techniques in classifying liver nodules in cirrhotic livers. The major diagnostic criteria are defined, and the sensitivity and specificity of each major diagnostic criteria are presented according to the literature. The existing differences between techniques in assessing the major diagnostic features can be then exploited in order to ensure a better classification and a better clinical management of liver nodules in cirrhotic livers. Ancillary features depend on the imaging technique used, and their presence can upgrade or downgrade the LI-RADS score of an observation, but only as far as LI-RADS 4. MRI is the imaging technique that provides the greatest number of ancillary features, whereas CEUS has fewer ancillary features than other imaging techniques. In the final part of the manuscript, some recommendations are made by the authors in order to guidephysicians as to when adding another imaging technique can be helpful in managing liver nodules in cirrhotic livers.
ABSTRACT
OBJECTIVES: To assess the multiparametric MRI (mpMRI) appearances of normal peripheral zone (PZ) across age groups in a biopsy-naïve population, where prostate cancer (PCa) was subsequently excluded, and propose a scoring system for background PZ changes. METHODS: This retrospective study included 175 consecutive biopsy-naïve patients (40-74 years) referred with a suspicion of PCa, but with subsequent negative investigations. Patients were grouped by age into categories ≤ 54, 55-59, 60-64, and ≥ 65 years. MpMRI sequences (T2-weighted imaging [T2WI], diffusion-weighted imaging [DWI]/apparent diffusion coefficient [ADC], and dynamic contrast-enhanced imaging [DCE]) were independently evaluated by two uro-radiologists on a proposed 4-point grading scale for background change on each sequence, wherein score 1 mirrored PIRADS-1 change and score 4 represented diffuse background change. Peripheral zone T2WI signal intensity and ADC values were also analyzed for trends relating to age. RESULTS: There was a negative correlation between age and assigned background PZ scores for each mpMRI sequence: T2WI: r = - 0.52, DWI: r = - 0.49, DCE: r = - 0.45, p < 0.001. Patients aged ≤ 54 years had mean scores of 3.0 (T2WI), 2.7 (DWI), and 3.1 (DCE), whilst patients ≥ 65 years had significantly lower mean scores of 1.7, 1.4, and 1.9, respectively. There was moderate inter-reader agreement for all scores (range κ = 0.43-0.58). Statistically significant positive correlations were found for age versus normalized T2WI signal intensity (r = 0.2, p = 0.009) and age versus ADC values (r = 0.33, p = 0.001). CONCLUSION: The normal PZ in younger patients (≤ 54 years) demonstrates significantly lower T2WI signal intensity, lower ADC values, and diffuse enhancement on DCE, which may hinder diagnostic interpretation in these patients. The proposed standardized PZ background scoring system may help convey the potential for diagnostic uncertainty to clinicians. KEY POINTS: ⢠Significant, positive correlations were found between increasing age and higher normalized T2-weighted signal intensity and mean ADC values of the prostatic peripheral zone. ⢠Younger men exhibit lower T2-weighted imaging signal intensity, lower ADC values, and diffuse enhancement on dynamic contrast-enhanced imaging, which may hinder MRI interpretation. ⢠A scoring system is proposed which aims towards a standardized assessment of the normal background PZ. This may help convey the potential for diagnostic uncertainty to clinicians.
Subject(s)
Multiparametric Magnetic Resonance Imaging , Prostatic Neoplasms , Aged , Diffusion Magnetic Resonance Imaging , Humans , Magnetic Resonance Imaging , Male , Prostatic Neoplasms/diagnostic imaging , Retrospective StudiesABSTRACT
OBJECTIVE: To evaluate the inter- and intraobserver agreement of COVID-RADS and CO-RADS reporting systems among differently experienced radiologists in a population with high estimated prevalence of COVID-19. METHODS AND MATERIALS: Chest CT scans of patients with clinically-epidemiologically diagnosed COVID-19 were retrieved from an open-source MosMedData data set, randomised, and independently assigned COVID-RADS and CO-RADS grades by an abdominal radiology fellow, thoracic imaging fellow and a consultant cardiothoracic radiologist. The inter- and intraobserver agreement of the two systems were assessed using the Fleiss' and Cohen's κ coefficients, respectively. RESULTS: A total of 200 studies were included in the analysis. Both systems demonstrated moderate interobserver agreement, with κ values of 0.51 [95% confidence interval (CI): 0.46-0.56] and 0.55 (95% CI: 0.50-0.59) for COVID-RADS and CO-RADS, respectively. When COVID-RADS and CO-RADS grades were dichotomised at cut-off values of 2B and 4 to evaluate the agreement between grades representing different levels of clinical suspicion for COVID-19, the interobserver agreement became substantial with κ values of 0.74 (95% CI: 0.66-0.82) for COVID-RADS and 0.73 (95% CI: 0.65-0.81) for CO-RADS. The median intraobserver agreement was considerably higher for CO-RADS reaching 0.81 (95% CI: 0.43-0.76) compared with 0.60 (95% CI: 0.43-0.76) of COVID-RADS. CONCLUSIONS: COVID-RADS and CO-RADS showed comparable interobserver agreement, which was moderate when grades were compared head-to-head and substantial when grades were dichotomised to better reflect the underlying levels of suspicion for COVID-19. The median intraobserver agreement of CO-RADS was, however, considerably higher compared with COVID-RADS. ADVANCES IN KNOWLEDGE: This paper provides a comprehensive review of the newly introduced COVID-19 chest CT reporting systems, which will help radiologists of all sub-specialties and experience levels make an informed decision on which system to use in their own practice.
ABSTRACT
To this date, it is a major oncological challenge to optimally diagnose, stage, and manage intrahepatic cholangiocarcinoma (ICC). Imaging can not only diagnose and stage ICC, but it can also guide management. Hence, imaging is indispensable in the management of ICC. In this article, we review the pathology, epidemiology, genetics, clinical presentation, staging, pathology, radiology, and treatment of ICC.
Subject(s)
Bile Duct Neoplasms , Cholangiocarcinoma , Liver Neoplasms , Bile Duct Neoplasms/diagnostic imaging , Bile Duct Neoplasms/pathology , Bile Ducts, Intrahepatic/diagnostic imaging , Bile Ducts, Intrahepatic/pathology , Cholangiocarcinoma/diagnostic imaging , Cholangiocarcinoma/pathology , Humans , Neoplasm StagingABSTRACT
Accurate staging of bladder cancer (BC) is critical, with local tumor staging directly influencing management decisions and affecting prognosis. However, clinical staging based on clinical examination, including cystoscopy and transurethral resection of bladder tumor (TURBT), often understages patients compared to final pathology at radical cystectomy and lymph node (LN) dissection, mainly due to underestimation of the depth of local invasion and the presence of LN metastasis. MRI has now become established as the modality of choice for the local staging of BC and can be additionally utilized for the assessment of regional LN involvement and tumor spread to the pelvic bones and upper urinary tract (UUT). The recent development of the Vesical Imaging-Reporting and Data System (VI-RADS) recommendations has led to further improvements in bladder MRI, enabling standardization of image acquisition and reporting. Multiparametric magnetic resonance imaging (mpMRI) incorporating morphological and functional imaging has been proven to further improve the accuracy of primary and recurrent tumor detection and local staging, and has shown promise in predicting tumor aggressiveness and monitoring response to therapy. These sequences can also be utilized to perform radiomics, which has shown encouraging initial results in predicting BC grade and local stage. In this article, the current state of evidence supporting MRI in local, regional, and distant staging in patients with BC is reviewed. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:649-667.
Subject(s)
Multiparametric Magnetic Resonance Imaging , Urinary Bladder Neoplasms , Humans , Magnetic Resonance Imaging , Neoplasm Recurrence, Local , Neoplasm Staging , Urinary Bladder Neoplasms/diagnostic imaging , Urinary Bladder Neoplasms/pathologyABSTRACT
Surgical flaps are commonly used for pelvic reconstruction in a subgroup of patients with locally advanced or recurrent anorectal and gynecologic malignancies and following complications of pelvic irradiation. Surgical scenarios where flaps may be placed include (but are not limited to) extended or radical abdominal perineal resection (APR) and total pelvic exenteration (PE). Surgical flaps in pelvic reconstruction serve several functions, including reducing dead space and providing structural support, facilitating wound closure and cosmetic appearance, enhancing the postsurgical healing process, protecting anastomoses and helping to prevent adhesions of organs and viscera to adjacent structures and the pelvic side wall. The most commonly used surgical flaps in pelvic reconstruction surgery include the VRAM (Vertical Rectus Abdominis Muscle), MRAM (Modified Rectus Abdominis Myocutaneous flap), gracilis, sartorius and omental flaps. Surgical flaps can be mistaken for recurrent or residual tumor by radiologists who are not familiar with the appearance or surgical methods of flap placement, since flaps may have a mass-like appearance on cross sectional imaging, including CT and MRI. Recurrent neoplasm may be difficult to differentiate from postoperative changes of flap placement and associated postsurgical anatomic distortion. This review article focuses on understanding the nuances of surgically placed pelvic flaps and identifying their normal and abnormal appearances on magnetic resonance imaging (MRI) along a time continuum. Postsurgical complications, including hematoma, postoperative fluid collections, infection, ischemia, and necrosis as well as tumor recurrence on the initial and follow-up magnetic resonance imaging are illustrated and discussed.
Subject(s)
Pelvic Exenteration , Plastic Surgery Procedures , Female , Humans , Magnetic Resonance Imaging , Neoplasm Recurrence, Local/surgery , Surgical FlapsABSTRACT
Ultrasound (US) is an attractive diagnostic approach to identify both common and uncommon nipple pathologies, such asduct ectasia, nipple abscess, nipple leiomyoma, nipple adenoma, fibroepithelial polyp, ductal carcinoma in situ (restrictedto nipple), invasive carcinoma, and Paget’s disease. US is the reliable first-line imaging technique to assess nipplepathologies. It is useful to identify and characterize nipple lesions. Additionally, we have presented the mammography andMRI outcomes correlated with histopathologic features for the relevant cases.
ABSTRACT
Natural orifice transluminal endoscopic surgery (NOTES) is a novel surgical approach, currently performed for an array of conditions. Endoscopic procedures offer significant benefits, including lower cost, no surgical incisions, and shorter hospital stays. These advantages align with the current trends in health care, namely a push for "cost-effective care." There are a multitude of health issues which are now being addressed by the endoscopic surgical approach, including peroral endoscopic myotomy (POEM), which is a relatively new technique used in the treatment of achalasia. Endoscopic treatment utilized for GERD includes transoral incisionless fundoplication. Endoscopic bariatric surgical procedures include intragastric balloon placement, endoscopic sleeve gastroplasty, and revision of prior bariatric procedures including Roux-en-Y gastric bypass and conventional gastric sleeve procedures. Endoscopic clips are routinely utilized for achieving hemostasis, treating iatrogenic gastric and bowel ulcerations and perforations and for the closure of enteric fistulization. Novel endoscopic procedures are now replacing conventional surgery due to their non-invasive nature, faster recovery and lower healthcare costs. Radiologists need to understand how these procedures are performed, as well as expected post-procedural imaging appearance and potential complications.
Subject(s)
Bariatric Surgery/methods , Endoscopy, Gastrointestinal/methods , Natural Orifice Endoscopic Surgery/methods , Radiologists , Gastric Balloon , Gastrointestinal Tract/diagnostic imaging , Gastrointestinal Tract/surgery , Humans , Tomography, X-Ray ComputedABSTRACT
Diffusion-weighted imaging (DWI) is an increasingly utilized sequence in the assessment of abdominal and pelvic lesions. Benign lesions containing hemorrhagic products, with conglomerates of tightly packed blood cells or fibers, can have restricted water diffusion on DWI and apparent diffusion coefficient maps. Such lesions can have restricted diffusion erroneously attributed to malignancy. This review illustrates benign hemorrhagic lesions displaying restricted diffusion, with histopathologic correlation in relevant cases.