Artificial Intelligence in BI-RADS Categorization of Breast Lesions on Ultrasound: Can We Omit Excessive Follow-ups and Biopsies?

RATIONALE AND OBJECTIVES: Artificial intelligence (AI) systems have been increasingly applied to breast ultrasonography. They are expected to decrease the workload of radiologists and to improve diagnostic accuracy. The aim of this study is to evaluate the performance of an AI system for the BI-RADS category assessment in breast masses detected on breast ultrasound. MATERIALS AND METHODS: A total of 715 masses detected in 530 patients were analyzed. Three breast imaging centers of the same institution and nine breast radiologists participated in this study. Ultrasound was performed by one radiologist who obtained two orthogonal views of each detected lesion. These images were retrospectively reviewed by a second radiologist blinded to the patient's clinical data. A commercial AI system evaluated images. The level of agreement between the AI system and the two radiologists and their diagnostic performance were calculated according to dichotomic BI-RADS category assessment. RESULTS: This study included 715 breast masses. Of these, 134 (18.75%) were malignant, and 581 (81.25%) were benign. In discriminating benign and probably benign from suspicious lesions, the agreement between AI and the first and second radiologists was moderate statistically. The sensitivity and specificity of radiologist 1, radiologist 2, and AI were calculated as 98.51% and 80.72%, 97.76% and 75.56%, and 98.51% and 65.40%, respectively. For radiologist 1, the positive predictive value (PPV) was 54.10%, the negative predictive value (NPV) was 99.58%, and the accuracy was 84.06%. Radiologist 2 achieved a PPV of 47.99%, NPV of 99.32%, and accuracy of 79.72%. The AI system exhibited a PPV of 39.64%, NPV of 99.48%, and accuracy of 71.61%. Notably, none of the lesions categorized as BI-RADS 2 by AI were malignant, while 2 of the lesions classified as BI-RADS 3 by AI were subsequently confirmed as malignant. By considering AI-assigned BI-RADS 2 as safe, we could potentially avoid 11% (18 out of 163) of benign lesion biopsies and 46.2% (110 out of 238) of follow-ups. CONCLUSION: AI proves effective in predicting malignancy. Integrating it into the clinical workflow has the potential to reduce unnecessary biopsies and short-term follow-ups, which, in turn, can contribute to sustainability in healthcare practices.

Does deep learning software improve the consistency and performance of radiologists with various levels of experience in assessing bi-parametric prostate MRI?

OBJECTIVE: To investigate whether commercially available deep learning (DL) software improves the Prostate Imaging-Reporting and Data System (PI-RADS) scoring consistency on bi-parametric MRI among radiologists with various levels of experience; to assess whether the DL software improves the performance of the radiologists in identifying clinically significant prostate cancer (csPCa). METHODS: We retrospectively enrolled consecutive men who underwent bi-parametric prostate MRI at a 3 T scanner due to suspicion of PCa. Four radiologists with 2, 3, 5, and > 20 years of experience evaluated the bi-parametric prostate MRI scans with and without the DL software. Whole-mount pathology or MRI/ultrasound fusion-guided biopsy was the reference. The area under the receiver operating curve (AUROC) was calculated for each radiologist with and without the DL software and compared using De Long's test. In addition, the inter-rater agreement was investigated using kappa statistics. RESULTS: In all, 153 men with a mean age of 63.59 ± 7.56 years (range 53-80) were enrolled in the study. In the study sample, 45 men (29.80%) had clinically significant PCa. During the reading with the DL software, the radiologists changed their initial scores in 1/153 (0.65%), 2/153 (1.3%), 0/153 (0%), and 3/153 (1.9%) of the patients, yielding no significant increase in the AUROC (p > 0.05). Fleiss' kappa scores among the radiologists were 0.39 and 0.40 with and without the DL software (p = 0.56). CONCLUSIONS: The commercially available DL software does not increase the consistency of the bi-parametric PI-RADS scoring or csPCa detection performance of radiologists with varying levels of experience.

A Comparative Study of Multiparametric MRI Sequences in Measuring Prostate Cancer Index Lesion Volume.

Objectives: To compare the effectiveness of individual multiparametric prostate MRI (mpMRI) sequences-T2W, diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC), and dynamic contrast-enhanced (DCE)-in assessing prostate cancer (PCa) index lesion volume using whole-mount pathology as the ground-truth; to assess the impact of an endorectal coil (ERC) on the measurements. Materials and Methods: We retrospectively enrolled 72 PCa patients who underwent 3T mpMRI with (n = 39) or without (n = 33) an ERC. A pathologist drew the index lesion borders on whole-mount pathology using planimetry (whole-mountvol). A radiologist drew the borders of the index lesion on each mpMRI sequence-T2Wvol, DWIvol, ADCvol, and DCEvol. Additionally, we calculated the maximum index lesion volume for each patient (maxMRIvol). The correlation and differences between mpMRI and whole-mount pathology in measuring the index lesion volume and the impact of an ERC were investigated. Results: The median T2Wvol, DWIvol, ADCvol, DCEvol, and maxMRIvol were 0.68 cm3, 0.97 cm3, 0.98 cm3, 0.82 cm3, and 1.13 cm3. There were good positive correlations between whole-mountvol and mpMRI sequences. However, all mpMRI-derived volumes underestimated the median whole-mountvol volume of 1.97 cm3 (P ≤ 0.001), with T2Wvol having the largest volumetric underestimation while DWIvol and ADCvol having the smallest. The mean relative index lesion volume underestimations of maxMRIvol were 39.16% ± 32.58% and 7.65% ± 51.91% with and without an ERC (P = 0.002). Conclusion: T2Wvol, DWIvol, ADCvol, DCEvol, and maxMRIvol substantially underestimate PCa index lesion volume compared with whole-mount pathology, with T2Wvol having the largest volume underestimation. Additionally, using an ERC exacerbates the volume underestimation.

Non-invasive Assessment of Subclinical Renal Parenchymal Changes in Chronic Hepatitis B Virus By T1 Mapping Magnetic Resonance Imaging.

Background: Renal parenchymal changes are seen in chronic hepatitis B virus (HBV) infection, and its disease diagnosis should be confirmed by renal biopsy, which is an invasive technique. Apparent-T1 mapping magnetic resonance imaging (MRI) is an established imaging technique that assesses subclinical tissue injury without using a contrast agent. Aims: To investigate the early stage subclinical renal changes without apparent renal dysfunction in patients with chronic HBV infection by renal apparent-T1 mapping MRI. Study Design: A cross-sectional study. Methods: This study included 45 participants with normal kidney function, wherein 25 have biopsy-proven chronic HBV hepatitis and 20 are healthy individuals. Liver and kidney biochemical tests were performed within 1 month before the MRI scan, and the estimated glomerular filtration rate was calculated by diet modification in renal disease formula. Breath-hold, electrocardiogram-gated Modified Look-Locker Imaging sequence was acquired in the coronal plane without contrast agent administration. Apparent-T1 mapping value was measured by manually drawing a region of interest in six points for both kidneys by two observers. Apparent-T1 mapping values were compared between the two groups. Results: The mean apparent-T1 mapping values of the kidneys were significantly higher in patients with chronic HBV infection compared to the control group (1445 ± 129 ms vs. 1306 ± 115 ms, P = 0.003). Inter-class correlation coefficient measurement analysis showed excellent agreement. Conclusion: Renal apparent-T1 mapping MRI may help show the early stage of renal parenchymal disease without apparent renal dysfunction in chronic HBV infection.

Liver parenchymal changes and association with cardiac magnetic resonance imaging findings in repaired tetralogy of Fallot: an intravoxel incoherent motion magnetic resonance imaging study.

BACKGROUND: Liver disease can develop in repaired tetralogy of Fallot (TOF) from hepatic congestion caused by volume and pressure overload of the right ventricle. Noninvasive assessment of the liver is important for diagnosing and managing children with TOF. OBJECTIVE: To evaluate subclinical hepatic changes without liver function test abnormality in adolescents with repaired TOF using intravoxel incoherent motion (IVIM) MRI and cardiac MRI findings. MATERIALS AND METHODS: We included 106 young adults (75 with repaired TOF and 31 healthy individuals) in the study. Liver IVIM MRI examinations were performed with 10 b values (0, 10, 20, 30, 50, 80, 100, 200, 400, 800 s/mm2). Two observers measured IVIM MRI parameters D true, D* and f, as well as apparent diffusion coefficient (ADC) values in liver segments 5-8. RESULTS: D* and f values were significantly lower in adolescents with TOF (P = 0.003 vs. P = 0.05, respectively). ADC values were higher in adolescents with TOF (P = 0.005). However, we found no significant difference between adolescents with and without TOF in terms of Dtrue (P = 0.53). There was a significant correlation between f value and right ventricular ejection fraction. The intraclass correlation coefficient (ICC) analysis of the two observers showed substantial-to-excellent agreement for D, f, D true and ADC (0.7, 0.8, 0.9 and 0.8, respectively). CONCLUSION: The results of our study suggest that impaired microperfusion with increased ADC values in adolescents with repaired TOF reflect hepatic congestion rather than fibrosis. Hepatic congestion characterized by decreased ADC values can be easily differentiated before fibrotic changes occur by using IVIM MRI to assess diffusion and microcapillary perfusion separately.

Cardiometabolic risk factors in Turkish children with hepatosteatosis.

Aslan A, Erdemli S, Durukan Günaydin G, Aslan M, Yazar RÖ, Kabaalioglu A, Agirbasli MA. Cardiometabolic risk factors in Turkish children with hepatosteatosis. Turk J Pediatr 2019; 61: 714-722. We aimed to investigate the prevalence of cardiometabolic (CM) risk factors (impaired fasting glucose ( > 100 mg/dL), high blood pressure, overweight or obesity, high serum triglycerides (TG) and low serum high-density lipoprotein cholesterol levels) in children with hepatosteatosis detected by abdominal ultrasound. Children whom ultrasound examination revealed hepatic steatosis were included in the study. Medical records, anthropometric and biochemical parameters were reviewed for the presence of the CM risk factors. Presence of ≥3 risk factors was defined as metabolic syndrome (MS). One hundred and forty-eight children and adolescents (67 boys, 81girls, and mean age 12.1±2.7 years) with hepatosteatosis were included. Sonographic hepatosteatosis grades of 1, 2 and 3 were observed in 111 (75%), 33 (22.3%), and 4 (2.7%) subjects, respectively. MS was observed in 36 patients (24.3%). The number of CM risk factors and degree of hepatic steatosis were correlated (r=0.183, p=0.026). Serum TG levels in girls and age in boys were significantly associated with the presence of medium to severe hepatosteatosis (grades 2 or 3) (R2=.342, =.040 and R2=.538, p=.001, respectively). CM risk factors and MS are common in children with hepatosteatosis. The presence and grade of hepatosteatosis on ultrasound can be used as surrogate markers of MS and CM risk in children.

Assessment of Peripheral Nerves With Shear Wave Elastography in Type 1 Diabetic Adolescents Without Diabetic Peripheral Neuropathy.

OBJECTIVES: To investigate the utility of shear wave elastography (SWE) in detecting morphologic abnormalities of the median nerve and posterior tibial nerve in transverse and longitudinal axes in adolescents with type 1 diabetes mellitus (DM) without diabetic peripheral neuropathy (DPN). METHODS: The median nerves and posterior tibial nerves of 25 adolescents with diagnosis and follow-up of type 1 DM without DPN and 32 healthy volunteers were evaluated with SWE by 2 observers on the transverse and longitudinal axes. The cross-sectional area and thickness of the nerves and disease duration were noted, and probable associations of these parameters with SWE features were analyzed. Interobserver and intraobserver correlations were also examined. The statistical significance level was set at P < .05. RESULTS: Both the median nerve and posterior tibial nerve were smaller, thinner, and stiffer in the patient group for both observers on both axes. The disease duration weakly correlated with median nerve SWE features (r = 0.245-0391). The thickness and cross-sectional area had no correlations with SWE features. CONCLUSIONS: The median nerve and posterior tibial nerve in adolescents with type 1 DM without DPN have morphologic abnormalities that can be displayed by SWE regardless of the imaging axis. Shear wave elastography may have a potential role in subclinical DPN, but the reliability of the findings is not as high as desirable.