Abdominal Radiology最新文献

The evaluation of patients with gastrointestinal (GI) bleeding is complicated due to the variety of tests available and the large number of potential causes of bleeding. MRI is less commonly used than computed tomography and endoscopy but it can diagnose disease that causes GI bleeding and serve as a complementary role to other tests. MRI is most often used in the form of magnetic resonance enterography (MRE) to assess patients with suspected bleeding from the small bowel. While CT enterography (CTE) and video capsule endoscopy (VCE) are the more commonly used tests in the setting of GI Bleeding, MRE has characteristics which may make it the more favorable modality for a given patient. Potential advantages of MRE, protocol considerations and the literature delineating its diagnostic performance for detecting pathology which can cause GI bleeding relative to CTE and VCE are reviewed here. MRI is uncommonly used to assess patients with upper GI bleeding, lower GI bleeding and patients with bleeding sites that remain undetected despite a formal evaluation of the GI tract, however it may add value in specific clinical scenarios. These uncommon scenarios and specific clinical examples are also presented to highlight the potential benefits of MRI.

Purpose

To evaluate radiological and pathological factors associated with mesorectal lymph node (MLN) metastases in prostate cancer patients, as detected by Prostate Specific Membrane Antigen (PSMA) imaging with Positron Emission Tomography/Computed Tomography (PET/CT).

Methods

In a retrospective case-control study of 135 prostate cancer patients who underwent PSMA PET/CT between January 2022 and October 2024, forty-five patients with positive mesorectal nodal uptake were matched 1:2 with control patients based on age and International Society of Urological Pathology (ISUP) grade. All patients had multiparametric prostate MR imaging and biopsy results available. We analyzed radiological features (PI-RADS scores, extraprostatic extension, seminal vesicle invasion) and histopathological characteristics (ISUP grade, percentage of positive biopsy cores, cribriform growth, perineural invasion) using univariable and multivariable logistic regression.

Results

In multivariable analysis, two independent factors associated with mesorectal lymph node involvement were identified: presence of non-mesorectal lymph node metastases (OR 32.86, CI: 7.10-152.03, p < 0.001) and larger prostate volume (OR 2.00, CI: 1.14–3.51, p = 0.016). Patients with positive MLN uptake had higher prostate specific antigen (PSA) levels (19.60 vs. 8.23 ng/mL, p = 0.02) and more frequent seminal vesicle invasion (44.4% vs. 26.7%, p = 0.04) in univariable analysis, but these associations were not significant in the multivariable model.

Conclusion

Non-mesorectal lymph node metastases and increased prostate volume are independently associated with mesorectal lymph node involvement in prostate cancer. These findings suggest that careful evaluation of the mesorectal region should be performed in patients with non-mesorectal regional lymph node metastases or larger prostate volumes, potentially impacting treatment planning and radiation field design.

Purpose

Radiology reports are essential for communicating imaging findings to guide diagnosis and treatment. Although most radiology reports are accurate, errors can occur in the final reports due to high workloads, use of dictation software, and human error. Advanced artificial intelligence models, such as GPT-4, show potential as tools to improve report accuracy. This retrospective study evaluated how GPT-4 performed in detecting and correcting errors in finalized radiology reports in real-world settings for abdominopelvic computed tomography (CT) reports.

Methods

We evaluated finalized CT abdominopelvic reports from a tertiary health system by using GPT-4 with zero-shot learning techniques. Six radiologists each reviewed 100 of their finalized reports (randomly selected), evaluating GPT-4’s suggested revisions for agreement, acceptance, and clinical impact. The radiologists’ responses were compared by years in practice and sex.

Results

GPT-4 identified issues and suggested revisions for 91% of the 600 reports; most revisions addressed grammar (74%). The radiologists agreed with 27% of the revisions and accepted 23%. Most revisions were rated as having no (44%) or low (46%) clinical impact. Potential harm was rare (8%), with only 2 cases of potentially severe harm. Radiologists with less experience (≤ 7 years of practice) were more likely to agree with the revisions suggested by GPT-4 than those with more experience (34% vs. 20%, P = .003) and accepted a greater percentage of the revisions (32% vs. 15%, P = .003).

Conclusions

Although GPT-4 showed promise in identifying errors and improving the clarity of finalized radiology reports, most errors were categorized as minor, with no or low clinical impact. Collectively, the radiologists accepted 23% of the suggested revisions in their finalized reports. This study highlights the potential of GPT-4 as a prospective tool for radiology reporting, with further refinement needed for consistent use in clinical practice.

Artificial intelligence (AI) algorithms and deep learning continue to change the landscape of radiology. New algorithms promise to enhance diagnostic accuracy, improve workflow efficiency, and automate repetitive tasks. This article provides a narrative review of the FDA-cleared AI algorithms which are commercially available in the United States as of late 2024 and targeted toward assessment of abdominopelvic organs and related diseases, evaluates potential advantages of using AI, and suggests future directions for the field.

Emphysematous conditions of the abdomen and pelvis are uncommon but are potentially lethal. A majority of these share common predisposing risk factors such as uncontrolled diabetes mellitus, immunocompromised states, chronic kidney disease, hepatic cirrhosis, malignancy, chemotherapy, and polytrauma. Computed tomography is the most utilized modality for the evaluation of emphysematous conditions in an emergency scenario. It is equally important to recognize mimics of the lethal emphysematous pathologies to avoid misdiagnosis. Gas-containing imaging mimics occur due to recent interventions, instrumentation, and fistulae. This review discusses the pertinent clinical and imaging features of emphysematous conditions of the abdomen and pelvis that can help confirm the diagnosis and guide patient management. For each of these emphysematous pathologies, we also discuss various gas-containing imaging mimics, as the management of these mimics differs from true emphysematous pathologies.

Graphical abstract

Advancements in prostate imaging, particularly multiparametric magnetic resonance imaging (mpMRI), have transformed the diagnosis and monitoring paradigms for prostate cancer (PCa). However, considerable interobserver variability has been reported,emphasizing the importance of standardized scoring and reporting systems. The Prostate Imaging Reporting and Data System (PI-RADS), along with quality assurance protocols such as Prostate Imaging Quality (PI-QUAL), has enhanced the diagnostic accuracy of mpMRI. While mpMRI effectively identifies extraprostatic extension, emerging hybrid positron emission tomography/magnetic resonance imaging (PET/MRI) and radio labeled prostate-specific membrane antigen (PSMA) tracers have improved metastasis detection. In active surveillance settings, the Prostate Cancer Radiologic Estimation of Change in Sequential Evaluation (PRECISE) recommendations enable the assessment of longitudinal PCa changes. Post-treatment evaluation can also employ dedicated frameworks, such as the Prostate Imaging for Recurrence Reporting (PI-RR) after radical prostatectomy, and either the Prostate Imaging after Focal Ablation (PI-FAB) or the Transatlantic Recommendations for Prostate Gland Evaluation with MRI after Focal Therapy (TARGET). This article reviews key aspects of mpMRI, including structured interpretation frameworks beyond the widely recognized PI-RADS.

Endometriosis, affecting approximately 10% of reproductive-age women, presents significant diagnostic challenges due to its variable clinical presentation and anatomic distributions. Despite advances in imaging technology, accurate detection and characterization of endometriotic lesions remain complex, with diagnostic delays averaging 7–12 years from symptom onset. This review analyzes the spectrum of imaging strengths and potential pitfalls across imaging modalities used to assess endometriosis focusing on technical factors and anatomic blind spots that can impact diagnostic accuracy and clinical management. Strengths of ultrasound include dynamic capabilities and real-time assessment with patient feedback, though it requires experienced operators and standardized protocols. MRI offers comprehensive evaluation with excellent soft tissue contrast and multiplanar capabilities but faces challenges with static imaging and artifact interference. CT, while limited by radiation exposure and intermediate contrast resolution, shows promise in specific scenarios, particularly evaluation of endometriosis deposits on the bowel. This review details optimal patient preparation strategies, including the role of vaginal and rectal gel opacification, bowel preparation, and contrast administration. Special attention is given to challenging anatomic areas including thoracic, gastrointestinal, abdominal wall, urinary tract and neurovascular involvement. The manuscript emphasizes the importance of standardized reporting templates for communication of findings to facilitate surgical planning. A multidisciplinary approach to imaging patients with known deep invasive endometriosis helps ensure the best modality is chosen for the clinical question at hand.