Diagnostic and interventional radiology最新文献

Purpose: To assess the impact of a commercially available computed tomography (CT)-based artificial intelligence (AI) software for detecting acute intracranial hemorrhage (AIH) on radiologists' diagnostic performance and workflow in a real-world clinical setting.

Methods: This retrospective study included a total of 956 non-contrast brain CT scans obtained over a 70-day period, interpreted independently by 2 board-certified general radiologists. Of these, 541 scans were interpreted during the initial 35 days before the implementation of AI software, and the remaining 415 scans were interpreted during the subsequent 35 days, with reference to AIH probability scores generated by the software. To assess the software's impact on radiologists' performance in detecting AIH, performance before and after implementation was compared. Additionally, to evaluate the software's effect on radiologists' workflow, Kendall's Tau was used to assess the correlation between the daily chronological order of CT scans and the radiologists' reading order before and after implementation. The early diagnosis rate for AIH (defined as the proportion of AIH cases read within the first quartile by radiologists) and the median reading order of AIH cases were also compared before and after implementation.

Results: A total of 956 initial CT scans from 956 patients [mean age: 63.14 ± 18.41 years; male patients: 447 (47%)] were included. There were no significant differences in accuracy [from 0.99 (95% confidence interval: 0.99-1.00) to 0.99 (0.98-1.00), P = 0.343], sensitivity [from 1.00 (0.99-1.00) to 1.00 (0.99-1.00), P = 0.859], or specificity [from 1.00 (0.99-1.00) to 0.99 (0.97-1.00), P = 0.252] following the implementation of the AI software. However, the daily correlation between the chronological order of CT scans and the radiologists' reading order significantly decreased [Kendall's Tau, from 0.61 (0.48-0.73) to 0.01 (0.00-0.26), P < 0.001]. Additionally, the early diagnosis rate significantly increased [from 0.49 (0.34-0.63) to 0.76 (0.60-0.93), P = 0.013], and the daily median reading order of AIH cases significantly decreased [from 7.25 (Q1-Q3: 3-10.75) to 1.5 (1-3), P < 0.001] after the implementation.

Conclusion: After the implementation of CT-based AI software for detecting AIH, the radiologists' daily reading order was considerably reprioritized to allow more rapid interpretation of AIH cases without compromising diagnostic performance in a real-world clinical setting.

Clinical significance: With the increasing number of CT scans and the growing burden on radiologists, optimizing the workflow for diagnosing AIH through CT-based AI software integration may enhance the prompt and efficient treatment of patients with AIH.

Purpose: This study aimed to evaluate and compare the diagnostic performance of various Thyroid Imaging Reporting and Data Systems (TIRADS), with a particular focus on the artificial intelligence-based TIRADS (AI-TIRADS), in characterizing thyroid nodules.

Methods: In this retrospective study conducted between April 2016 and May 2022, 1,322 thyroid nodules from 1,139 patients with confirmed cytopathological diagnoses were included. Each nodule was assessed using TIRADS classifications defined by the American College of Radiology (ACR-TIRADS), the American Thyroid Association (ATA-TIRADS), the European Thyroid Association (EU-TIRADS), the Korean Thyroid Association (K-TIRADS), and the AI-TIRADS. Three radiologists independently evaluated the ultrasound (US) characteristics of the nodules using all classification systems. Diagnostic performance was assessed using sensitivity, specificity, positive predictive value (PPV), and negative predictive value, and comparisons were made using the McNemar test.

Results: Among the nodules, 846 (64%) were benign, 299 (22.6%) were of intermediate risk, and 147 (11.1%) were malignant. The AI-TIRADS demonstrated a PPV of 21.2% and a specificity of 53.6%, outperforming the other systems in specificity without compromising sensitivity. The specificities of the ACR-TIRADS, the ATA-TIRADS, the EU-TIRADS, and the K-TIRADS were 44.6%, 39.3%, 40.1%, and 40.1%, respectively (all pairwise comparisons with the AI-TIRADS: P < 0.001). The PPVs for the ACR-TIRADS, the ATA-TIRADS, the EU-TIRADS, and the K-TIRADS were 18.5%, 17.9%, 17.9%, and 17.4%, respectively (all pairwise comparisons with the AI-TIRADS, excluding the ACR-TIRADS: P < 0.05).

Conclusion: The AI-TIRADS shows promise in improving diagnostic specificity and reducing unnecessary biopsies in thyroid nodule assessment while maintaining high sensitivity. The findings suggest that the AI-TIRADS may enhance risk stratification, leading to better patient management. Additionally, the study found that the presence of multiple suspicious US features markedly increases the risk of malignancy, whereas isolated features do not substantially elevate the risk.

Clinical significance: The AI-TIRADS can enhance thyroid nodule risk stratification by improving diagnostic specificity and reducing unnecessary biopsies, potentially leading to more efficient patient management and better utilization of healthcare resources.

Purpose: To quantitatively evaluate the vascularity of the thyroid parenchyma in patients diagnosed with Hashimoto's thyroiditis (HT) compared with healthy controls by using vascularity index (VI) through power Doppler (PD) and color superb microvascular imaging (cSMI) and to determine a threshold VI value to effectively differentiate patients with HT and hypothyroid HT.

Methods: This prospective cross-sectional study involved 73 patients diagnosed with HT and 66 healthy controls. The diagnosis of HT was established based on clinical and laboratory findings. The total volume of the thyroid gland was measured, and the region of interest was drawn manually by delineating the gland boundaries for VI calculation on PD and cSMI. The mean VI for both lobes of the thyroid were computed for each participant. Statistical analyses were conducted using SPSS version 29.0, with receiver operating characteristic curve analysis employed to ascertain the optimal cSMI VI cut-off values for the diagnosis of HT and for patients with hypothyroid HT.

Results: The analysis revealed no significant differences in the total thyroid volume between the HT group and the control group, or between the hypothyroid and euthyroid HT subgroups. The SMI VI values were recorded at 8.85 [interquartile range (IQR): 25%-75%, 6.55-12.6] for patients with HT and 8.40 (IQR: 25%-75%, 6.70-12.8) for the control group, indicating a statistically significant increase in the HT cohort (P < 0.001). Additionally, the PD VI values in patients with HT were significantly higher than in the control group (P < 0.001). A strong positive correlation was identified between thyroid-stimulating hormone levels and cSMI VI in patients with HT (rho = 0.739, P < 0.001), whereas the correlation with PD VI was found to be weak (rho = 0.346, P < 0.001). The optimal cut-off value for SMI VI was 6.75% for the general diagnosis of HT and 8.825% for patients with hypothyroid HT.

Conclusion: This study indicates that the optimal threshold values of 6.75% for the diagnosis of HT and 8.825% for patients with hypothyroid HT suggest that cSMI is an effective and promising diagnostic tool for detecting alterations in thyroid vascularization. Furthermore, there is a strong concordance among radiologists regarding the VI measurements.

Clinical significance: The SMI technique represents a promising diagnostic tool for the detection of subtle alterations in thyroid vascularization. The higher sensitivity of cSMI in comparison to PD positions it as an innovative and effective technology for the assessment of HT, offering valuable insights into disease activity and progression.

Purpose: This study investigates the competence of a newly certified radiologist in reporting hydrops imaging and examines the role of magnetic resonance imaging (MRI) findings in diagnosing definite and probable Ménière's disease (MD).

Methods: Sixty-four cases were retrospectively evaluated-blinded to clinical data-by a senior radiologist (O-1) and a newly certified radiologist (O-2) using 3D heavily T2-weighted and delayed contrast-enhanced three-dimensional fluid-attenuated inversion recovery sequences. The posterior fossa-posterior semicircular canal (P-P) distance, endolymphatic hydrops (EH), perilymphatic enhancement (PE), and the round window sign (RWS) were assessed.

Results: Interobserver agreement was moderate for cochlear (κ = 0.591) and vestibular hydrops (κ = 0.566), good for PE (κ = 0.663), and excellent for the RWS (κ = 0.817). O-1 demonstrated good intraobserver agreement for the RWS (κ = 0.787) and excellent agreement for the other parameters. O-2 showed lower intraobserver agreement for cochlear hydrops, vestibular hydrops, and the RWS (κ = 0.366, κ = 0.332, and κ = 0.398, respectively). The P-P distance showed excellent interobserver [intraclass correlation coefficient (ICC) = 0.932] and intraobserver agreement (ICC = 0.978 for O-1; ICC = 0.886 for O-2). The P-P distance was significantly shorter in definite MD (dMD) than in probable MD (pMD) (1.23 ± 1.07 mm vs. 2.17 ± 1.79 mm, P = 0.021). The rate and grade of hydrops were higher in dMD (P < 0.050), whereas the RWS was more frequent in pMD. Hydrops and PE were more often observed on the symptomatic side (P < 0.001). Cochlear hydrops was identified in 14.3% and vestibular hydrops in 31.2% of asymptomatic sides.

Conclusion: The newly certified radiologist's intraobserver agreement for hydrops imaging was insufficient. In dMD, the retrolabyrinthine bone is thinner, hydrops is more frequent and advanced, and the RWS is less common. Approximately one in five patients with MD may have a perilymphatic fistula. Close monitoring of asymptomatic contralateral ears is essential.

Clinical significance: Accurate MRI evaluation of EH in MD strongly depends on the radiologist's expertise. This study highlights that newly certified radiologists may show lower reliability in assessing hydrops imaging, underscoring the need for targeted training programs.

Purpose: To investigate the association between patellofemoral malalignment and early-stage trochlear and patellar chondromalacia using the T2* mapping method.

Methods: Seventy-five patients were included in the study and divided into two groups based on the presence (patient group) or absence (control group) of patellofemoral malalignment on magnetic resonance imaging. The T2* mapping measurements were evaluated by dividing patellar and trochlear cartilage into 12 quadrants on sagittal slices. The groups were first compared based on the mean T2* relaxation times of the cartilage. Subsequently, the 12 quadrants were compared individually between the two groups. Cut-off values were calculated for the quadrants, with significant differences observed.

Results: The patient group included 39 patients, and the control group included 36 patients. There was no significant difference between the groups in terms of mean T2* relaxation values for the trochlear and patellar cartilage. However, in the separate comparison of the 12 quadrants, T2* relaxation values in the upper-outer-outer (P1, T1) and upper-outer-inner (P2, T2) quadrants of both the trochlear and patellar cartilage were found to be statistically significantly higher in the patient group. Similarly, significant cut-off values were identified for the T1, P1, and P2 quadrants.

Conclusion: Early chondromalacia can be quantitatively detected using T2* mapping. In patients with elevated T2* relaxation values in the superior-lateral regions of the patellar and trochlear cartilage, patellofemoral malalignment should be considered in the etiology.

Clinical significance: Chondromalacia caused by patellofemoral malalignment may exhibit an asymmetric onset, with the superior-lateral quadrant as the initial site of cartilage damage in both trochlear and patellar cartilage.

Purpose: This study aimed to compare the value of differences (Δ) in parameters obtained via both dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) and flourine-18 fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) between baseline and post-neoadjuvant therapy in predicting the pathological response to neoadjuvant therapy in breast cancer.

Methods: A total of 109 patients who underwent both baseline and post-neoadjuvant therapy DCE-MRI and ¹⁸F-FDG PET/CT examinations were retrospectively analyzed. The DCE-MRI parameters and ¹⁸F-FDG PET/CT parameters [metabolic tumor volume (MTV), standardized uptake value (SUV)_max, SUV_mean, and total lesion glycolysis] were recorded at both time points. Additionally, the Δs between these parameters were calculated. Postsurgical pathology reports were documented, and the patients were subsequently categorized into two groups: those exhibiting pathologic complete response (pCR) and those exhibiting partial response. Parameters from DCE-MRI and ¹⁸F-FDG PET/CT were compared to determine which predicted pathological response to neoadjuvant therapy more effectively.

Results: Patients with partial response demonstrated a higher rate of histologic grade 3 than those with pCR (P = 0.030). The only DCE-MRI parameter to indicate a significant difference between the two groups (P = 0.024) was the Δ(%)wash-out rate. Among the baseline parameters, only MTV successfully predicted pathological response (P = 0.033). The only post-neoadjuvant therapy parameter to be predictive of pathological response (P = 0.003) was SUV_mean. In receiver operating characteristic analysis, ΔSUV_mean emerged as the most significant parameter for predicting pathological response, followed by post-neoadjuvant SUV_mean [area under the curve: 0.724 (95% confidence interval: 0.630-0.805) and 0.673 (0.577-0.760), respectively].

Conclusion: The Δ¹⁸F-FDG PET/CT parameters are better than ΔDCE-MRI in predicting pathologic response to neoadjuvant therapy. Among these parameters, ΔSUV_mean is the most successful.

Clinical significance: Neoadjuvant chemotherapy (NAC) response is one of the most important criteria in breast cancer prognosis. The two most important imaging modalities in breast cancer diagnosis and follow-up protocols are MRI and ¹⁸F-FDG PET/CT. However, it is not clear which of these two modalities is more successful in predicting the difference in treatment response between baseline and post-NAC.