Background The 2023 International Federation of Gynecology and Obstetrics (FIGO) staging system for endometrial cancer (EC) integrates histopathologic and molecular variables that may be inapparent at imaging. Purpose To evaluate the impact of the updated FIGO 2023 staging system on preoperative MRI assessment in patients with EC, and to assess its impact on patient outcomes. Materials and Methods In this dual-center retrospective study, women with biopsy-proven EC underwent abdominal MRI. Radiologic FIGO (rFIGO) 2009 stage was assigned from MRI and biopsy grade and histotype. Surgical specimens provided pathologic FIGO (pFIGO) 2009 and 2023 stages, incorporating molecular status. Concordance and causes of discordance were assessed. Potential impact on initial management was inferred using 2025 European Society of Gynaecological Oncology-European Society for Radiotherapy and Oncology-European Society of Pathology guidelines. Disease-free survival (DFS) and overall survival (OS) were assessed with Kaplan-Meier. Results A total of 231 women (median age, 63 years; IQR, 55-73 years) were evaluated. Major-stage agreement decreased from 86.6% (200 of 231) of patients for rFIGO 2009 versus pFIGO 2009 to 76.2% (176 of 231) for rFIGO 2023 versus pFIGO 2023 (P < .001); substage concordance declined from 74.0% (171 of 231) with FIGO 2009 to 56.3% (130 of 231) with FIGO 2023 (P < .001). Under FIGO 2023, the most common causes of discordance were minimal myometrial invasion (37 of 101 women, 36.6%), lymph node involvement (18 of 101 women, 17.8%), and histology-dependent factors such as lymphovascular space invasion (12 of 101 women, 11.9%) and histologic changes (19 of 101 women, 18.8%). Molecular data reclassified 35 of the 231 (15.2%) patients (18 patients were downstaged; 17 were upstaged). Discrepancies would have altered the initial approach in 22 of the 231 (9.5%) patients. Recurrence and death occurred in 18 (7.8%) and 16 (7.0%) of the 231 patients, respectively, with DFS and/or OS stratification across radiologic and pathologic 2009 and 2023 stages (all P < .001). Conclusion The 2023 FIGO update reduced substage concordance between preoperative pelvic MRI and final pathology, primarily due to the missed detection of minimal myometrial invasion and micrometastatic lymph node involvement. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Kataoka and Himoto in this issue.

Purpose. High-grade serous ovarian carcinoma (HGSOC) is characterized by pronounced biological and spatial heterogeneity and is frequently diagnosed at an advanced stage. Neoadjuvant chemotherapy (NACT) followed by delayed primary surgery is commonly employed in patients unsuitable for primary cytoreduction. The Chemotherapy Response Score (CRS) is a validated histopathological biomarker of response to NACT, but it is only available postoperatively. In this study, we investigate whether pre-treatment computed tomography (CT) imaging and clinical data can be used to predict CRS as an investigational decision-support adjunct to inform multidisciplinary team (MDT) discussions regarding expected treatment response. Methods. We proposed a 2.5D multimodal deep learning framework that processes lesion-dense omental slices using a pre-trained Vision Transformer encoder and integrates the resulting visual representations with clinical variables through an intermediate fusion module to predict CRS. Results. Our multimodal model, integrating imaging and clinical data, achieved a ROC-AUC of 0.95 alongside 95

Gynaecological tumours present a broad spectrum of histological subtypes due to the diverse anatomical and tissue origin of the reproductive organs. Rare tumours affect less than 6 per 100 000 individuals annually, posing significant challenges in diagnosis and management due to limited clinical awareness. Indeed, treatment protocols rely on options developed for more common histotypes, which may have limited efficacy on these rare tumours. In recent years, collaborative international efforts have started to address these gaps, improving standards of care. A comprehensive understanding of rare tumours' clinical and imaging features is necessary for radiologists in order to provide clinicians with useful information for treatment planning. In this review, we adopted an organ-based outline, describing rare tumours of the uterine corpus (leiomyosarcoma, endometrial stromal sarcoma, carcinosarcoma), cervix (gastric-type adenocarcinoma), and ovary (cystadenofibroma, lipid-poor teratoma, struma ovarii, immature teratoma, dysgerminoma). Additionally, tumours occurring at multiple sites, including lymphoma, neuroendocrine tumours, aggressive angiomyxoma and metastases are discussed. The objective is to help radiologists become familiar with these uncommon entities, ultimately increasing awareness on this topic.

Decidual changes in deep pelvic endometriosis (DPE) are extremely rare. During pregnancy, endometriotic activity and symptoms diminish due to hormonal changes, whereas the endometrial lining undergoes decidualization in response to progesterone. In some cases, a significant hormonal boost can promote the decidualization of endometriotic implants, with endometriomas being the most common sites for such changes. Decidualized endometriosis at sites outside the ovaries is a rarer phenomenon, and the associated imaging features have been studied less. We present a case of non-pregnancy-related decidualization involving multiple sites of DPE. A 33-year-old woman exposed to ovarian stimulation before in vitro fertilization (IVF) presented with chronic pelvic pain and vaginal bleeding. Emergency CT demonstrated irregular solid tissue in the pouch of Douglas and along the bowel loops, raising concern for a neoplastic process. Tumor markers were negative. Transvaginal ultrasound showed a hypoechoic, irregular, hypervascular lesion, and ultrasound-guided biopsy demonstrated decidualized stromal endometriosis. MRI confirmed a solid mass centered in the posterior cul-de-sac with intermediate T2-weighted (W) signal intensity, hemorrhagic foci on fat-suppressed T1W images, marked diffusion restriction, and avid post-contrast enhancement. Similar solid components were also present at multiple pelvic sites, adjacent to DPE implants. Short-interval MRI follow-up showed size reduction; however, due to persistent pain, fertility-sparing surgery was performed, and histology confirmed decidualization of DPE at all sites. This rare case emphasizes that decidualization can also occur on DPE and outside pregnancy, potentially mimicking malignant transformation, highlighting the role of MRI in aiding diagnosis and guiding proper management for these patients.

Objectives: High-grade serous ovarian carcinoma (HGSOC) is typically diagnosed at an advanced stage with extensive peritoneal metastases, making treatment challenging. Neoadjuvant chemotherapy (NACT) is often used to reduce tumor burden before surgery, but about 40

BackgroundHigh-grade serous carcinoma is a highly metastatic disease with a limited longterm disease control from systemic anti-cancer treatment, for which the radiological treatment response assessment metrics are imprecise. In this work, we developed noninvasive imagingbased measurements of spatial and longitudinal heterogeneity in a retrospective analysis of a phase 2 non-randomized study of germline BRCA1/BRCA2 mutated (gBRCAm) ovarian cancer patients treated with combination of PARP inhibitors (PARPi) and immune checkpoint inhibitors (ICIs).MethodsLesions identified in CT images at baseline, week 4 (after PARPi only) and week 12 (after 8 weeks of PARPi + ICIs) were manually segmented. Anatomical networks of the metastatic sites were constructed to represent patterns of disease distribution. Volume and first-order radiomic features were computed and compared to different assessments of treatment response.ResultsThe average number of edges per patient in the anatomical networks and total volumetric burden decreased with treatment were measured, differentiating between responders and nonresponders. Changes in volume at week 4 provided better indication of long-term response than the default RECIST assessment at the same time-point. Significant differences were also found between responders and non-responders in the first-order radiomic feature Energy.ConclusionsIn this feasibility study, we have demonstrated that noninvasive image-based analysis can identify quantitative imaging features associated with the response to the combination of PARPi and ICIs. These can be used to identify markers of response to ICIs from negative trials of a disease with limited response to ICIs.

To evaluate the impact of MRI-defined bladder wall invasion from uterine cervical cancer (CC) on disease recurrence and overall survival. IRB-approved multicenter retrospective study including women who underwent staging MRI for histologically confirmed CC (Jan 2015–Dec 2020). Image analysis was independently performed by two radiologists. Bladder wall invasion was diagnosed if ≥ 3 of the following criteria were met: loss of the cervix-bladder fat plane, bladder wall thickening, loss of bladder wall T2-hypointensity, and presence of endoluminal tumor growth. MRI findings were compared with endoscopy/cytology. The impact of MRI-defined bladder wall invasion on tumor recurrence and survival was assessed using logistic regression. Survival curves were compared using the log-rank test. We included 214 women with a median age of 55 (IQR 47–65) years. MRI-defined bladder wall invasion was observed in 21.5

The goal is to investigate the best time point for assessing radiological complete response after exclusive chemoradiation in locally advanced cervical cancer (LACC). This is a retrospective single-center study. Seventy-nine patients with LACC, stage IB3-IVA FIGO 2018 treated between January and December 2020 were retrospectively analyzed. All patients received external beam radiotherapy (45 Gy in 25 daily fractions ± simultaneous boost to lymph nodes), and interventional radiotherapy (IRT, 28 Gy/twice/weekly) with concurrent chemotherapy. The radiological complete response evaluation was examined using magnetic resonance imaging (MRI) at three timepoints: (i) before IRT, at the end of external beam radiotherapy, (ii) 3 months following the completion of IRT and (iii) 6 months after IRT. Seventy-nine patients were included. At the three timepoints, the complete response rate increased with 21, 53, and 59 patients reporting a complete response at MRI scan, respectively. Seven patients with partial response at the second assessment had complete response 6 months after treatment completion, overall resulting in 80

Imaging is used for lymphoma detection, Ann Arbor/Lugano staging, and treatment response assessment. [18F]FDG PET/CT should be used for most lymphomas, including Hodgkin lymphoma, aggressive/high-grade Non-Hodgkin lymphomas (NHL) such as diffuse large B-cell lymphoma, and many indolent/low-grade NHLs such as follicular lymphoma. Apart from these routinely FDG-avid lymphomas, some indolent NHLs, such as marginal zone lymphoma, are variably FDG-avid; here, [18F]FDG PET/CT is an alternative to contrast-enhanced CT at baseline and may be used for treatment response assessment if the lymphoma was FDG-avid at baseline. Only small lymphocytic lymphoma/chronic lymphocytic leukemia (SLL/CLL) should exclusively undergo CT at baseline and follow-up unless transformation to high-grade lymphoma is suspected. While [18F]FDG PET/CT is sufficient to rule out bone marrow involvement in Hodgkin lymphoma, biopsy may be needed in other lymphomas. The 5-point (Deauville) score for [18F]FDG PET that uses the liver and blood pool uptake as references should be used to assess treatment response in all FDG-avid lymphomas; post-treatment FDG uptake ≤ liver uptake is considered complete response. In all other lymphomas, CT should be used to determine changes in lesion size; for complete response, resolution of all extranodal manifestations, and for lymph nodes, long-axis decrease to ≤ 1.5 cm are required.

To summarize the key updates introduced in the 2023 International Federation of Gynecology and Obstetrics (FIGO) classification for endometrial cancer (EC), and to highlight the role of MRI in aligning with these changes for improved staging and patient management. A review of the updated 2023 FIGO classification, which integrates molecular profiling and histopathological criteria, was conducted. Additionally, the revised European Society of Urogenital Radiology (ESUR) MRI recommendations were analyzed to assess their alignment with the new FIGO framework, focusing on their role in evaluating myometrial invasion (MI) and cervical stromal involvement. The updated FIGO classification incorporates molecular data to refine risk stratification and staging accuracy. MRI continues to play a pivotal role in distinguishing between stages, mapping disease extent, and guiding surgical planning. The updated ESUR recommendations emphasize standardized MRI protocols, particularly the use of multiphase contrast-enhanced imaging, to improve diagnostic confidence in assessing MI. The integration of molecular classification into FIGO staging, supported by standardized and advanced MRI protocols as recommended by ESUR, enhances the management of endometrial cancer. Question The 2023 FIGO update integrates molecular profiling into endometrial cancer staging, requiring MRI adaptations to improve accuracy in assessing disease extent, including myometrial invasion. Findings Updated ESUR MRI guidelines emphasize multiphase contrast-enhanced imaging, structured reporting, and integration with FIGO 2023 classification, enhancing diagnostic precision for staging and treatment planning. Clinical relevance Standardized MRI protocols aligned with FIGO 2023 system improve endometrial cancer staging, guiding optimal surgical and therapeutic strategies, reducing diagnostic variability, and enhancing patient outcomes through individualized risk stratification and personalized treatment.

This study presents an investigation of the potential of radiomic features extracted from postmortem computed tomography (PMCT) scans of the lungs to provide valuable insights into the postmortem interval (PMI), a crucial parameter in forensic medicine. Sequential PMCT scans were performed on 17 bodies with known times of death, ranging from 4 to 108 h postmortem. Radiomic features were extracted from the lungs, and a mixed-effects model, tailored for sequential data, was employed to assess the relationship between feature values and the PMI. Four model variants were tested to identify the most suitable functional form for describing this association. Several statistically significant trends between the PMI and radiomic features were observed, with twelve distinct features demonstrating selective relevance to postmortem changes in the lungs. Notably, cluster shade, a grey-level co-occurrence matrix (GLCM) feature, significantly decreased with the PMI, the median intensity increased over time, and the root mean squared feature values tended to decrease. The retained features included first-order statistical metrics, shape-based characteristics, and second-order texture attributes, which may reflect alterations such as gas formation and structural modifications within the lungs. This study highlights the potential of PMCT scan-based radiomics as a complementary tool to enhance existing postmortem interval estimation methods. These findings reinforce the role of quantitative imaging techniques in forensic investigations.

PURPOSE:: High-grade serous ovarian carcinoma (HGSOC) is characterised by significant spatial and temporal heterogeneity, often presenting at an advanced metastatic stage. One of the most common treatment approaches involves neoadjuvant chemotherapy (NACT), followed by surgery. However, the multi-scale complexity of HGSOC poses a major challenge in evaluating response to NACT. METHODS:: Here, we present a multi-task deep learning approach that facilitates simultaneous segmentation of pelvic/ovarian and omental lesions in contrast-enhanced computerised tomography (CE-CT) scans, as well as treatment response assessment in metastatic ovarian cancer. The model combines multi-scale feature representations from two identical U-Net architectures, allowing for an in-depth comparison of CE-CT scans acquired before and after treatment. The network was trained using 198 CE-CT images of 99 ovarian cancer patients for predicting segmentation masks and evaluating treatment response. RESULTS:: It achieves an AUC of 0.78 (95% CI [0.70-0.91]) in an independent cohort of 98 scans of 49 ovarian cancer patients from a different institution. In addition to the classification performance, the segmentation Dice scores are only slightly lower than the current state-of-the-art for HGSOC segmentation. CONCLUSION:: This work is the first to demonstrate the feasibility of a multi-task deep learning approach in assessing chemotherapy-induced tumour changes across the main disease burden of patients with complex multi-site HGSOC, which could be used for treatment response evaluation and disease monitoring.

Radiology is one of the medical specialties most significantly impacted by Artificial Intelligence (AI). AI systems, particularly those employing machine and deep learning, excel in processing large datasets and comparing images from similar contexts, fulfilling radiological demands. However, the implementation of AI in radiology presents notable challenges, including concerns about data privacy, informed consent, and the potential for external interferences affecting decision-making processes. Biases represent another critical issue, often stemming from unrepresentative datasets or inadequate system training, which can lead to distorted outcomes and exacerbate healthcare inequalities. Additionally, generative AI systems may produce 'hallucinations' arising from their reliance on probabilistic modeling without the ability to distinguish between true and false information. Such risks raise ethical and legal questions, especially when AI-induced errors harm patient health. Concerning liability for medical errors involving AI, healthcare professionals currently retain full accountability for their decisions. AI systems remain tools to support, not replace, human expertise and judgment. Nevertheless, the "black box" nature of many AI models - wherein the reasoning behind outputs remains opaque - limits the possibility of fully informed consent. We advocate for prioritizing Explainable Artificial Intelligence (XAI) in radiology. While potentially less performant than black-box models, XAI enhances transparency, allowing patients to understand how their data is used and how AI influences clinical decisions, aligning with ethical standards.

Background: Currently, the landscape of AI-driven chatbots is growing and proposing new and extraordinarily beneficial solutions to common problems. In addition to powerful chatbots such as ChatGPT, Bing Chat, and Google Bard, which are wide-ranging applications capable of performing multiple different tasks, some applications focus on specific functionality to improve the user experience. One product that helps with document reading experiences is ChatPDF, which enables AIs to read, extract data, and reply to user inquiries. Our objective is to test this AI tool to improve data extraction from clinical articles. Methods: We assessed AI comprehension in 48 diverse scientific articles, examining sections such as the main topic, conclusions, results, materials and methods, sampling and data collection, and additional information from attached images, graphs, and tables. Five healthcare professionals, who possess expertise in emergency/abdominal radiology, thoracic radiology, musculoskeletal radiology, neuroradiology, and endocrinology, participated in an evaluation study. The primary objective was to assess answers using a revised Mean Opinion Score (MOS) scale (0 to 5 points) and compare performance across distinct article sections using the Gini heterogeneity index. Results: The finding demonstrated an adequate comprehension of the topic and the conclusions, a poor ability to extract information from the images, graphs, and tables attached to the article, and an insufficient performance regarding the methods of sampling and data collection. Conclusions: ChatPDF can be useful to extract principal information from PDF articles, but not further in-depth details. An accurate and thorough guide to their use is essential, considering the potential and critical issues already raised about these artificial intelligence (AI) systems. Ours is an attempt to explore the topic further, which certainly requires validation.

PURPOSE:To assess the incidence of pelvic insufficiency fractures (PIFs) after concurrent chemoradiotherapy (CCRT) in patients with locally advanced cervical cancer (LACC), their time of onset and risk factors. We also analysed the inter-observer agreement between gynaecologic radiologists (GYN readers) and radiologists expert in musculoskeletal imaging (MSK reader) in detecting PIFs in our tertiary care centre. METHODS:Patients with confirmed LACC who underwent concurrent chemoradiation (CCRT) at our institution from June 2019 to November 2022 were retrospectively included. These patients underwent follow-up pelvic MRI every 3-6 months after CCRT. Cohen's kappa statistics was employed to assess the inter-observer agreement between GYN and MSK readers.Logistic regression analysis was performed calculating odds ratios (OR) to identify risk factors for PIFs, such as age, body mass index (BMI), diabetes, smoking, hypertension, renal function and tumour size. RESULTS:Eighty-seven patients were included. PIFs were diagnosed in 21/87 (24.1 %) patients with a median onset time of 7.4 months from the end of EBRT. Among risk factors, age was statistically associated with PIFs (OR = 1.057, 95 % CI: 1.005-1.118, p = 0.033) with median age in the fracture group of 61.1 years (range: 52.0-71.5) and 53.8 years (range: 43.8-63.3). BMI was a significant predictor of PIFs (OR = 1.134; 95 % CI: 1.013-1.285; p = 0.027), with a higher median BMI among patients with PIFs (26.5; range: 21.5-31.2) compared to non-fractured patients (23.1; range: 20.2-25.1). Also patients with reduced renal function (eGFR < 60 mL/min) had 3.437 times higher odds of experiencing fractures compared to those with normal eGFR. The GYN readers correctly identified PIFs in 2/21 cases and agreed with the MSK reader in 68/87 cases. The interobserver agreement was poor to fair (K = 0.138; 95 % CI: 0-0.311). CONCLUSIONS:PIFs are a common complication of CCRT. Their identification on post-CCRT MRI may decrease the need for further targeted investigations and invasive treatments.

In recent years, post-mortem imaging has advanced with techniques such as Post-Mortem Computed Tomography (PMCT) and Post-Mortem Magnetic Resonance imaging (PMMR). PMCT is particularly useful for assessing skeletal injuries, vascular lesions, and estimating post-mortem intervals. These analyses are based on radiomics principles, a growing field that requires specialized expertise, which is still limited. To support this field, postmortem imaging biobanks are proposed as valuable tools, especially when enhanced with Artificial Intelligence (AI). However, complexities in this field require caution and adherence to guidelines, especially regarding consent for image storage, participant information, international cooperation, and nonprofit operations.

To evaluate the capability of hyperpolarized [1-13C] pyruvate MRI to predict pathologic response to neoadjuvant treatment in multi-site abdominopelvic disease of high-grade serous ovarian cancer (HGSOC) patients and to compare 13C MRI and [18F]-FDG PET/CT measurements for detecting early treatment response. We recruited eight patients with HGSOC in this prospective study who underwent 13C MRI and [18F]-FDG PET/CT before and after the first cycle of neoadjuvant chemotherapy treatment (NACT). Imaging parameters were compared with clinical and histophatologic parameters. We demonstrate here that 13C MRI of hyperpolarized [1-13C]pyruvate metabolism in multiple abdominal metastases resulted in rapid labeling of the endogenous tumor lactate pool. The rate of labeling was similar between the different anatomical disease sites and independent of tumor volume. The apparent rate constant describing exchange of 13C label between pyruvate and lactate (kPL) was positively correlated with PET standard uptake values (SUVmax) for [18F]-FDG in metastatic tumor deposits in the ovary/pelvis (R = 0.471, P = 0.02). Decreased lactate labeling could be detected after the first cycle of neoadjuvant chemotherapy and was associated with pathological response. There was no overall decrease in lactate labeling in a single patient who lacked a complete histopathologic response. kPL was associated with cancer tissue LDHA concentration (rho = 0.641; P = 0.02). This exploratory study demonstrates the potential of 13C MRI measurements for assessing early response to neoadjuvant chemotherapy in patients with HGSOC.