Diagnostic test accuracy study of a commercially available deep learning algorithm for ischemic lesion detection on brain MRIs in suspected stroke patients from a non-comprehensive stroke center.

PURPOSE: To estimate the ability of a commercially available artificial intelligence (AI) tool to detect acute brain ischemia on Magnetic Resonance Imaging (MRI), compared to an experienced neuroradiologist. METHODS: We retrospectively included 1030 patients with brain MRI, suspected of stroke from January 6th, 2020 to 1st of April 2022, based on these criteria: Age ≥ 18 years, symptoms within four weeks before the scan. The neuroradiologist reinterpreted the MRI scans and subclassified ischemic lesions for reference. We excluded scans with interpretation difficulties due to artifacts or missing sequences. Four MRI scanner models from the same vendor were used. The first 800 patients were included consecutively, remaining enriched for less frequent lesions. The index test was a CE-approved AI tool (Apollo version 2.1.1 by Cerebriu). RESULTS: The final analysis cohort comprised 995 patients (mean age 69 years, 53 % female). A case-based analysis for detecting acute ischemic lesions showed a sensitivity of 89 % (95 % CI: 85 %-91 %) and specificity of 90 % (95 % CI: 87 %-92 %). We found no significant difference in sensitivity or specificity based on sex, age, or comorbidities. Specificity was reduced in cases with DWI artifacts. Multivariate analysis showed that increasing ischemic lesion size and fragmented lesions were independently associated with higher sensitivity, while non-acute lesion ages lowered sensitivity. CONCLUSIONS: The AI tool exhibits high sensitivity and specificity in detecting acute ischemic lesions on MRI compared to an experienced neuroradiologist. While sensitivity depends on the ischemic lesions' characteristics, specificity depends on the image quality.

Diagnosis of bladder tumours in patients with macroscopic haematuria: a prospective comparison of split-bolus computed tomography urography, magnetic resonance urography and flexible cystoscopy.

OBJECTIVE: The aim of this study was to compare split-bolus computed tomography urography (CTU), magnetic resonance urography (MRU) and flexible cystoscopy in patients with macroscopic haematuria regarding the diagnosis of bladder tumours. MATERIALS AND METHODS: In this prospective study, 150 patients underwent CTU, MRU and flexible cystoscopy. Two uroradiologists individually reviewed the images without any clinical information, using a questionnaire. Patient records and pathology reports were also reviewed. RESULTS: At flexible cystoscopy, MRU and CTU, 32, 19 and 15 bladder lesions were identified, respectively. Histopathology showed that 13 of the 29 biopsied lesions were transitional cell carcinomas. Compared with the histopathology, the sensitivity and specificity for detection of tumours by CTU and MRU were 61.5% and 94.9%, and 79.9% and 93.4%, respectively. False-positive detection of bladder tumours, compared with histopathology, was reported in seven CTUs and nine MRUs, whereas the number of false-negative findings was five for CTUs and three for MRUs. CONCLUSIONS: Split-bolus CTU or MRU cannot replace cystoscopy in cases of macroscopic haematuria. MRU has a higher sensitivity than split-bolus CTU, and is a potential alternative to flexible cystoscopy. Patients with a low risk of bladder cancer may forgo flexible cystoscopy if a bladder tumour is identified by either CTU or MRU, and proceed straight to transuretheral resection of the bladder.

Upper urinary tract tumors: how does the contrast enhancement measured in a split-bolus CTU correlate to histological staging?

BACKGROUND: Computed tomography urography (CTU) is used widely in the work-up of patients with symptoms of urinary tract lesions. Preoperative knowledge of whether a tumor is invasive or non-invasive is important for the choice of surgery. So far there are no studies about the distinction of invasive and non-invasive tumors in ureter and renal pelvis based on the enhancement measured with Hounsfield Units. PURPOSE: To examine the value of CTU using split-bolus technique to distinguish non-invasive from invasive urothelial carcinomas in the upper urinary tract. MATERIAL AND METHODS: Patients who underwent nephroureterectomy between 2006 and 2011 and who had split-bolus CTU prior to surgery were included. The images were available electronically. The attenuation values before and after administration of iodine-based contrast media were measured. The radiology, patient, and pathology records were reviewed. RESULTS: Of the 158 patients who underwent unilateral nephroureterectomy 69 fulfilled the inclusion criteria. Histopathological examination revealed 31 patients with non-invasive and 38 with invasive urothelial carcinoma. Neither absolute attenuation nor change in attenuation values obtained at CTU could distinguish between invasive and non-invasive lesions. No patients had a CTU within the last year before the examination that resulted in surgery. CONCLUSION: A split-bolus CTU cannot distinguish between invasive and non-invasive urothelial tumors in the upper urinary tract, but the examination is useful to diagnose a tumor in the renal pelvis and the ureter.