European Radiology最新文献

Objectives: To develop an automated pipeline for extracting prostate cancer-related information from clinical notes.

Materials and methods: This retrospective study included 23,225 patients who underwent prostate MRI between 2017 and 2022. Cancer risk factors (family history of cancer and digital rectal exam findings), pre-MRI prostate pathology, and treatment history of prostate cancer were extracted from free-text clinical notes in English as binary or multi-class classification tasks. Any sentence containing pre-defined keywords was extracted from clinical notes within one year before the MRI. After manually creating sentence-level datasets with ground truth, Bidirectional Encoder Representations from Transformers (BERT)-based sentence-level models were fine-tuned using the extracted sentence as input and the category as output. The patient-level output was determined by compilation of multiple sentence-level outputs using tree-based models. Sentence-level classification performance was evaluated using the area under the receiver operating characteristic curve (AUC) on 15% of the sentence-level dataset (sentence-level test set). The patient-level classification performance was evaluated on the patient-level test set created by radiologists by reviewing the clinical notes of 603 patients. Accuracy and sensitivity were compared between the pipeline and radiologists.

Results: Sentence-level AUCs were ≥ 0.94. The pipeline showed higher patient-level sensitivity for extracting cancer risk factors (e.g., family history of prostate cancer, 96.5% vs. 77.9%, p < 0.001), but lower accuracy in classifying pre-MRI prostate pathology (92.5% vs. 95.9%, p = 0.002) and treatment history of prostate cancer (95.5% vs. 97.7%, p = 0.03) than radiologists, respectively.

Conclusion: The proposed pipeline showed promising performance, especially for extracting cancer risk factors from patient's clinical notes.

Clinical relevance statement: The natural language processing pipeline showed a higher sensitivity for extracting prostate cancer risk factors than radiologists and may help efficiently gather relevant text information when interpreting prostate MRI.

Key points: When interpreting prostate MRI, it is necessary to extract prostate cancer-related information from clinical notes. This pipeline extracted the presence of prostate cancer risk factors with higher sensitivity than radiologists. Natural language processing may help radiologists efficiently gather relevant prostate cancer-related text information.

Objectives: Risk calculators (RCs) improve patient selection for prostate biopsy with clinical/demographic information, recently with prostate MRI using the prostate imaging reporting and data system (PI-RADS). Fully-automated deep learning (DL) analyzes MRI data independently, and has been shown to be on par with clinical radiologists, but has yet to be incorporated into RCs. The goal of this study is to re-assess the diagnostic quality of RCs, the impact of replacing PI-RADS with DL predictions, and potential performance gains by adding DL besides PI-RADS.

Material and methods: One thousand six hundred twenty-seven consecutive examinations from 2014 to 2021 were included in this retrospective single-center study, including 517 exams withheld for RC testing. Board-certified radiologists assessed PI-RADS during clinical routine, then systematic and MRI/Ultrasound-fusion biopsies provided histopathological ground truth for significant prostate cancer (sPC). nnUNet-based DL ensembles were trained on biparametric MRI predicting the presence of sPC lesions (UNet-probability) and a PI-RADS-analogous five-point scale (UNet-Likert). Previously published RCs were validated as is; with PI-RADS substituted by UNet-Likert (UNet-Likert-substituted RC); and with both UNet-probability and PI-RADS (UNet-probability-extended RC). Together with a newly fitted RC using clinical data, PI-RADS and UNet-probability, existing RCs were compared by receiver-operating characteristics, calibration, and decision-curve analysis.

Results: Diagnostic performance remained stable for UNet-Likert-substituted RCs. DL contained complementary diagnostic information to PI-RADS. The newly-fitted RC spared 49% [252/517] of biopsies while maintaining the negative predictive value (94%), compared to PI-RADS ≥ 4 cut-off which spared 37% [190/517] (p < 0.001).

Conclusions: Incorporating DL as an independent diagnostic marker for RCs can improve patient stratification before biopsy, as there is complementary information in DL features and clinical PI-RADS assessment.

Clinical relevance statement: For patients with positive prostate screening results, a comprehensive diagnostic workup, including prostate MRI, DL analysis, and individual classification using nomograms can identify patients with minimal prostate cancer risk, as they benefit less from the more invasive biopsy procedure.

Key points: The current MRI-based nomograms result in many negative prostate biopsies. The addition of DL to nomograms with clinical data and PI-RADS improves patient stratification before biopsy. Fully automatic DL can be substituted for PI-RADS without sacrificing the quality of nomogram predictions. Prostate nomograms show cancer detection ability comparable to previous validation studies while being suitable for the addition of DL analysis.

Objectives: To compare the safety and efficiency of ultrasound-guided percutaneous radiofrequency ablation (RFA) and surgical resection (SR) for thyroid papillary carcinoma (PTC) in the danger triangle area.

Methods: The clinical data of 298 patients who underwent either percutaneous RFA or SR for PTC in the thyroid danger triangle at our hospital between January 2018 and April 2020 were retrospectively analyzed. Propensity score matching is employed to regulate for confounding factors. All patients undergoing ablation were treated using a strategy that combined sufficient paratracheal fluid isolation with a low-power, short electrode. Disease progression was analyzed in patients with T1N0M0 PTC (T1a and T1b) employed in Kaplan‒Meier curves. Treatment parameters and the rates of local recurrence, distant metastasis, and complications are recorded and compared.

Results: Of 182 eligible patients who were included, 91 were in the RFA (age 44.84 ± 13.19; 71 females; 77 T1a) and 91 were in the SR (age 47.36 ± 11.05; 68 females; 69 T1a). The average treatment time, length of hospital stays, blood loss volume, and scar length are substantially less in the RFA than in the SR. Major complications as well as postoperative permanent recurrent laryngeal nerve injury and postoperative transient parathyroid dysfunction occurred only in the SR, with a substantial distinction between the two groups (p < 0.05). There is no substantial distinction in the disease progression between RFA and SR treatment of T1N0M0 PTC.

Conclusion: RFA is as effective as surgery for PTC in the danger triangle area in the short term, with faster recovery and fewer complications.

Clinical relevance statement: Radiofrequency ablation has a clinical efficacy comparable to surgery in the treatment of papillary thyroid carcinoma in the danger triangle area in the short term with the advantages of faster recovery and fewer complications when compared with surgery.

Key points: Use of radiofrequency ablation (RFA) in the thyroid danger triangle is still controversial. RFA and surgery groups showed no difference in disease progressions, and no major complications occurred with RFA. Radiofrequency ablation offers a new option for papillary thyroid carcinoma patients in the danger triangle.

Objectives: Our aim was twofold. First, to validate Anali scores with and without gadolinium (ANALI_Gd and ANALI_NoGd) in primary sclerosing cholangitis (PSC) patients. Second, to compare the ANALIs prognostic ability with the recently-proposed potential functional stricture (PFS).

Materials and methods: This retrospective study included 123 patients with a mean age of 41.5 years, who underwent gadoxetic acid-enahnced MRI (GA-MRI). Five readers independently evaluated all images for calculation of ANALI_Gd and ANALI_NoGd scores based upon following criteria: intrahepatic bile duct change severity, hepatic dysmorphia, liver parenchymal heterogeneity, and portal hypertension. In addition, hepatobiliary contrast excretion into first-order bile ducts was evaluated on 20-minute hepatobiliary-phase (HBP) images to assess PFS. Inter- and intrareader agreement were calculated (Fleiss´and Cohen kappas). Kaplan-Meier curves were generated for survival analysis. ANALI_NoGd, ANALI_Gd, and PFS were correlated with clinical scores, labs and outcomes (Cox regression analysis).

Results: Inter-reader agreement was almost perfect (ϰ = 0.81) for PFS, but only moderate-(ϰ = 0.55) for binary ANALI_NoGd. For binary ANALI_Gd, the agreement was slightly better on HBP (ϰ = 0.64) than on arterial-phase (AP) (ϰ = 0.53). Univariate Cox regression showed that outcomes for decompensated cirrhosis, orthotopic liver transplantation or death significantly correlated with PFS (HR (hazard ratio) = 3.15, p < 0.001), ANALI_NoGd (HR = 6.42, p < 0.001), ANALI_GdHBP (HR = 3.66, p < 0.001) and ANALI_GdAP (HR = 3.79, p < 0.001). Multivariate analysis identified the PFS, all three ANALI scores, and Revised Mayo Risk Score as independent risk factors for outcomes (HR 3.12, p < 0.001; 6.12, p < 0.001; 3.56, p < 0.001;3.59, p < 0.001; and 4.13, p < 0.001, respectively).

Conclusion: ANALI_NoGd and GA-MRI-derived ANALI scores and PFS could noninvasively predict outcomes in PSC patients.

Clinical relevance statement: The combined use of Anali scores and the potential functional stricture (PFS), both derived from unenhanced-, and gadoxetic acid enhanced-MRI, could be applied as a diagnostic and prognostic imaging surrogate for counselling and monitoring primary sclerosing cholangitis patients.

Key points: Primary sclerosing cholangitis patients require radiological monitoring to assess disease stability and for the presence and type of complications. A contrast-enhanced MRI algorithm based on potential functional stricture and ANALI scores risk-stratified these patients. Unenhanced ANALI score had a high negative predictive value, indicating some primary sclerosing cholangitis patients can undergo non-contrast MRI surveillance.

Introduction: To investigate the relationship between collaterals and blood-brain barrier (BBB) permeability on pre-treatment MRI in a cohort of acute ischemic stroke (AIS) patients treated with thrombectomy.

Methods: We conducted a retrospective analysis of the HIBISCUS-STROKE cohort, a single-center observational study that enrolled patients treated with thrombectomy from 2016 to 2022. Dynamic-susceptibility MRIs were post-processed to generate K2 maps with arrival-time correction, which were co-registered with apparent diffusion coefficient (ADC) maps. The 90th percentile of K2 was extracted from the infarct core-defined by an ADC ≤ 620 × 10^-6 mm²/s with manual adjustments-and expressed as a percentage change compared to the contralateral white matter. Collaterals were assessed using pre-thrombectomy digital subtraction arteriography with an ASITN/SIR score < 3 defining poor collaterals.

Results: Out of 249 enrolled, 101 (40.6%) were included (median age: 72.0 years, 52.5% of males, median NIHSS score at admission: 15.0). Patients with poor collaterals (n = 44) had worse NIHSS scores (median: 16.0 vs 13.0, p = 0.04), larger infarct core volumes (median: 43.7 mL vs 9.5 mL, p < 0.0001), and higher increases in K2 (median: 346.3% vs 152.7%, p = 0.003). They were less likely to achieve successful recanalization (21/44 vs 51/57, p < 0.0001) and experienced more frequent hemorrhagic transformation (16/44 vs 9/57, p = 0.03). On multiple variable analysis, poor collaterals were associated with larger infarct cores (odds ratio (OR) = 1.12, 95% confidence interval (CI): [1.07, 1.17], p < 0.0001) and higher increases in K2 (OR = 6.63, 95% CI: [2.19, 20.08], p = 0.001).

Conclusion: Poor collaterals are associated with larger infarct cores and increased BBB permeability at admission MRI.

Clinical relevance statement: Poor collaterals are associated with a larger infarct core and increased BBB permeability at admission MRI of AIS patients treated with thrombectomy. These findings may have translational interests for extending thrombolytic treatment eligibility and developing neuroprotective strategies.

Key points: In AIS, collaterals and BBB disruption have been both linked to hemorrhagic transformation. Poor collaterals were associated with larger ischemic cores and increased BBB permeability on pre-treatment MRI. These findings could contribute to hemorrhagic transformation risk stratification, thereby refining clinical decision-making for reperfusion therapies.

Objectives: To determine prevalence in the symptomatic population of dorsal mucoid cysts centered on dorsal capsuloscapholunate septum (DCSS) using high-field magnetic resonance imaging (MRI) for anatomoclinical and epidemiological correlations.

Materials and methods: This single-center retrospective study analyzed all 3-Tesla MRIs consecutively performed for painful wrists in 295 patients. Two blinded readers performed measurements. The protocol included T1 spin echo and 3D proton density sequences with fat saturation. Inter-observer reliability was assessed using kappa and intra-class correlation coefficients for cyst detection and volumetry, respectively. Disagreements concerning cyst detection were resolved by a consensus reading. Cyst size, relationship to extrinsic and scapholunate ligaments (SL), continuity of SL, minimum distance to the posterior interosseous nerve (PIN), cyst communication with joint, and anatomical classifications of cysts were analyzed. Correlation tests were performed to assess associations.

Results: Two-hundred ninety-five patients (mean age 39.6 +/- 15.6 (standard deviation), 161 males) were evaluated for detection of dorsal wrist cysts identified in 150/295. In this subgroup, the mean age was 38.7 years (15-75), the sex ratio of 0.6 (59% women), and the median volume cyst of 8.7 mm³ (0.52-2555). Cyst detection, volume, and major axis measurements showed very high agreement between observers, respectively, 0.89, 0.96, and 0.91. 42 patients had dorsal SL pain. A weak negative correlation was found between distance to PIN and dorsal SL pain (r = -0.2415; p < 0.05) and a weak positive correlation between Guérini's classification and dorsal SL pain (r = 0.2466; p < 0.05).

Conclusion: High-field MRI is the modality of choice for the detection, anatomical, and volumetric assessment of dorsal cysts. Preoperative assessment will be aided by the proposed revised anatomical classification.

Clinical relevance statement: High-field MRI is the modality of choice for the anatomical study of dorsal ganglion cysts. It allows the radiologist to accurately describe the anatomical relationships, size, and visibility of the pedicle, essential information for the surgeon's preoperative assessment.

Key points: Dorsal mucoid wrist ganglion is a condition for which prevalence remains to be determined. High-field MRI is a reproducible imaging modality for the detection and assessment of dorsal wrist cysts. High-field MRI has a key role in the preoperative management of dorsal mucoid cysts.