Accuracy and Efficacy of Artificial Intelligence-Derived Automatic Measurements of Transthoracic Echocardiography in Routine Clinical Practice.

Background: Transthoracic echocardiography (TTE) is the gold standard modality for evaluating cardiac morphology, function, and hemodynamics in clinical practice. While artificial intelligence (AI) is expected to contribute to improved accuracy and is being applied clinically, its impact on daily clinical practice has not been fully evaluated. Methods: We retrospectively examined 30 consecutive patients who underwent AI-equipped TTE at a single institution. All patients underwent manual and automatic measurements of TTE parameters using the AI-equipped TTE. Measurements were performed by three sonographers with varying experience levels: beginner, intermediate, and expert. Results: A comparison between the manual and automatic measurements assessed by the experts showed extremely high agreement in the left ventricular (LV) filling velocities (E wave: r = 0.998, A wave: r = 0.996; both p < 0.001). The automated measurements of LV end-diastolic and end-systolic diameters were slightly smaller (-2.41 mm and -1.19 mm) than the manual measurements, although without significant differences, and both methods showing high agreement (r = 0.942 and 0.977, both p < 0.001). However, LV wall thickness showed low agreement between the automated and manual measurements (septum: r = 0.670, posterior: r = 0.561; both p < 0.01), with automated measurements tending to be larger. Regarding interobserver variabilities, statistically significant agreement was observed among the measurements of expert, intermediate, and beginner sonographers for all the measurements. In terms of measurement time, automatic measurement significantly reduced measurement time compared to manual measurement (p < 0.001). Conclusions: This preliminary study confirms the accuracy and efficacy of AI-equipped TTE in routine clinical practice. A multicenter study with a larger sample size is warranted.

Additive value of the right parasternal view for the assessment of aortic stenosis.

BACKGROUND: Although Doppler evaluation using a multiplanar method is recommended to assess the severity of aortic stenosis (AS) with transthoracic echocardiography, evidence on the diagnostic significance of a non-apical method is limited. This study aimed to compare the use of the apical with the use of the right parasternal view (RPV) method to evaluate AS severity and to examine the diagnostic significance of performing the RPV method in addition to the apical method during the evaluation. METHODS: This retrospective observational study included 276 consecutive patients (mean age: 79 ± 10 years; women, 56%) with severe AS (aortic valve area [AVA] ≤1.0cm2 ). The severity of AS according to the apical method and that according to the RPV for all subjects were compared, and the significance of performing the RPV method in addition to the apical method was examined. Furthermore, we compared the concordance group, in which the apical and RPV methods indicated matching in severity, and the discordant group, in which the apical and RPV methods did not indicate matching severity. RESULTS: Peak velocity (Vmax ), mean pressure gradient (MG) were significantly higher and the AVA, AVAi, and Doppler velocity index (DVI) were significantly smaller when the RPV was added to the apical view. Performing the RPV method in addition to the apical method significantly decreased the number of low PG AS cases (MG < 40 mmHg) from 69.9% to 65.0% and it increased the number of very severe AS cases (Vmax ≥ 5 m/s) from 8.7% to 14.5%. Deviation of Doppler angle was significantly greater in the discordant group compared to the concordant group (22.5 ± .6 vs. 31.8 ± 1.7, p < .001). CONCLUSIONS: By performing the RPV method in addition to the apical method to determine AS severity, the diagnosis of AS to be resolved in approximately 10% of cases. These results suggest that AS severity may be underestimated by using the apical method alone.

The triangular cord ratio and the presence of a cystic lesion in the triangular cord. Suggested new ultrasound findings in the early diagnosis of Biliary Atresia.

BACKGROUND: In biliary atresia (BA), the ultrasonic triangular cord (TC) sign is positive at ≥ 3 mm, but sometimes there is BA even if it is ≤ 3 mm. For improving the ultrasonographic diagnosis, we have established a new evaluation, adding the ratio of the anterior/posterior thickness (TC ratio) in the hyperechoic area and the presence of a cystic lesion in the triangular cord (TCC). METHODS: We examined 24 cases of suspected BA who demonstrated acholic stools from 2006 to 2020. We retrospectively reviewed the timing of ultrasonographic diagnosis, the gallbladder diameter, gallbladder mucosal irregularity, the TC sign, TCC, and the TC ratio. RESULTS: In the BA group (n = 10) vs the Non-BA group (n = 14), the age at ultrasonography was 75 ± 41.7 vs. 81 ± 39.1 days (p = 0.72), the gallbladder diameter was 12.1 ± 9.7 vs. 24.2 ± 6.96 mm (p = 0.02), irregularity of gallbladder mucosa was 7 cases vs. 1 case (p < 0.01), and TC sign was 3.9 ± 1.3 vs. 2.0 ± 0.49 mm (p = 0.01), respectively. TCC was observed in 8/10 cases in the BA group and none in the Non-BA group (p < 0.01). TC ratio was 3.40 ± 0.68 (BA group) and 1.59 ± 0.41 (Non-BA group) (p < 0.01). CONCLUSION: The ultrasonic TC ratio improves the diagnostic accuracy of BA. TCC is a specific finding in the BA group.

The utility of superb microvascular imaging for the detection of deep vein thrombosis.

Superb microvascular imaging (SMI) reduces motion artifacts significantly and allows visualization of low-velocity blood flow in vessels; thus, thrombi are clearly demonstrated compared to conventional B-mode and color Doppler imaging. Ultrasonographic images of the lower extremities of a 69-year-old woman with acute ischemic stroke complicated with deep vein thrombosis (DVT) are presented. Because SMI allows visualization of low-velocity blood flow, echo contrast between the thrombus and the surrounding flow makes it possible to clearly visualize the thrombus. SMI is a useful method to evaluate DVT, especially early after thrombus formation when it is difficult to delineate size and length because of echolucency.

Sonographic characterization of 271 hepatic hemangiomas with typical appearance on CT imaging.

PURPOSE: Hepatic hemangioma is the most frequent benign solid tumor that requires differentiation from hepatic malignancy on ultrasonography. Useful ultrasound findings are therefore required for diagnosis. SUBJECTS AND METHODS: The following factors were investigated for 271 masses diagnosed as hepatic hemangioma by contrast-enhanced computed tomography (CT) in 188 patients: hepatic subsegment location, shape, maximum diameter, internal echo level, hyperechoic rim, posterior echoes, marginal hypoechoic band (halo), lateral shadow, blood flow signal in the central portion of the mass, and underlying liver disease. RESULTS: Hepatic masses were classified by internal echo pattern as homogeneous hyperechoic (35.0%), homogeneous hypoechoic (22.9%), isoechoic (5.2%), mixed hyperechoic (22.1%), or mixed hypoechoic (14.8%) masses. Twelve masses were isoechoic, and could be recognized by the hyperechoic rim. Posterior echo enhancement was present in 66 masses. There were no cases of posterior echo attenuation. A halo was present in six masses, with the underlying condition being fatty infiltrated liver in two masses; a hypoechoic region with a small amount of fat deposit was present around the mass, and this resembled a halo. This finding was termed a "pseudohalo." There were no masses in which a lateral shadow was observed. In terms of blood flow signals in the central portion of the mass, pulsatile flow was present in only one mass. CONCLUSIONS: Important ultrasonographic findings of hepatic hemangioma are characterized as the absence of lateral shadow (100%) and no attenuation of posterior echoes (100%), while the presence of a hyperechoic rim is useful for detecting isoechoic hemangioma.

The value of dynamic imaging for standardization of abdominal ultrasonography.

PURPOSE: Digital video recording of dynamic images is a potential way to improve the reproducibility of abdominal ultrasonography (US). Static US and dynamic US were compared using contrast-enhanced computed tomography (CE-CT) as a reference standard, and the value of stored video images was verified. METHODS: The subjects were 120 individuals who had undergone CE-CT ≤1 month before undergoing US. After static US, dynamic US was carried out according to the Scanning Procedure for Abdominal Ultrasonography described in this paper, and records of 30 separate scans were saved on digital video discs (DVD) as individual video files for interpretation. The findings from static US and dynamic US were compared. RESULTS: With respect to the abdominal US signs covered in this evaluation, dynamic US achieved a detection rate equivalent to or better than that of static US. Detection rates were significantly more frequent on dynamic US for: enlarged hepatic hilum lymph nodes (P < 0.001); gallbladder wall thickness (P < 0.01); cystic lesions of the liver, bright liver, and splenomegaly (P < 0.05). CONCLUSION: Dynamic US is a valuable tool because it enables repeated evaluation of target organs with playback speed adjustment.