Tradup, Donald J.

Objectives Uniformity artifacts caused by defective transducer elements or scanner malfunctions degrade diagnostic image quality. Traditional quality control (QC) methods, such as phantom testing and visual or quantitative image analysis, can be labor‐intensive and limited in test frequency. This study aims to develop a deep learning framework to detect uniformity artifacts and complement traditional QC. Methods Clinical median images were generated by aggregating co‐registered grayscale ultrasound images acquired by each transducer and computing median intensity values across the image stack. A pretrained ResNet‐18 model was fine‐tuned on a dataset consisting of clinical median images from linear and curvilinear transducers. The dataset was divided into training, validation, and testing subsets, ensuring no overlap between training and test transducers. To assess generalizability, the model was also evaluated on an independent test set of 396 phantom‐validated images from linear and curvilinear transducers. Results The model achieved 100% accuracy on the first dataset's test set. On the independent test set, it attained 87.4% accuracy with high sensitivity (0.84) and specificity (0.88), demonstrating robust generalization. Occlusion sensitivity maps confirmed the model's attention to uniformity artifact regions. Conclusion The deep learning framework using clinical median images demonstrated robust performance across several linear and curvilinear transducer models. It could be integrated into the clinical QC workflow by automating artifact detection in an effective and timely manner. In our practice, it can flag median images classified as artifact‐present, reducing human review time by approximately 80% while preserving detection accuracy.

AbstractPurposeUltrasound grayscale imaging preset optimization has often been qualitative and dependent upon vendor application specialists. This study aimed to propose a systematic approach for grayscale imaging preset optimization and apply the approach in a clinical abdominal scan setting.MethodsA six‐step approach was detailed including identification of clinical task, adjustment of basic parameters, fine‐tuning of advanced parameters, image performance metrics confirmation, clinical evaluation and data analysis, and implementation of new presets and monitoring of clinical usage. Its application in an abdominal scanning task was described for each step with phantoms, volunteers, and software tools.ResultsClinical image data analytics facilitated the understanding of the imaging task, relevant transducers, and target characteristics, in addition to specific requests from radiologists. Quantitative measurements were made on global image contrast and gray map function. In addition, clinically relevant phantoms and volunteer scans without and with acoustic distortion layers were involved to determine the new presets. Furthermore, phantom signal to noise ratio study and clinical evaluation using volunteers with different body habitus were utilized to confirm the superiority of the new presets. Quantitative clinical usage monitoring demonstrated successful implementation of the new presets.ConclusionsA systematic approach for grayscale imaging preset optimization has been proposed and successfully applied for a specific clinical task. This approach was designed to be generalizable and relatively flexible, which would facilitate movement away from previous qualitative and subjective approaches.

AbstractPurposeIt is unclear if a 3D transducer with the special design of mechanical swing or 2D array could provide acceptable 2D grayscale image quality for the general diagnosis purpose. The aim of this study is to compare the 2D image quality of a 3D intracavitary transducer with a conventional 2D intracavitary transducer using clinically relevant phantom experiments.MethodsAll measurements were performed on a GE Logiq E9 scanner with both a 2D (IC5‐9‐D) and a 3D (RIC5‐9‐D) transducer used in 2D mode. Selection of phantom targets and acquisition parameters were determined from analysis of 33 clinical pelvic exams. Depth of penetration (DOP), contrast response, contrast of anechoic cylinders (diameter: 6.7 mm) at 1.5 and 4.5 cm depths in transverse planes, and in‐plane resolution represented by full‐width half‐maximum of pin targets at multiple depths were measured with transmit frequencies of 7 and 8 MHz. Spherical signal‐noise‐ratio (SNR) (diameter: 4 and 2 mm) at multiple depths were measured at 8 MHz.ResultsRIC5‐9‐D demonstrated <8% decrease in DOP for both transmit frequencies (7 MHz: 69.7 ± 8.2 mm; 8 MHz: 64.3 ± 7.8 mm) compared with those from IC5‐9‐D (7 MHz: 73.9 ± 4.4 mm; 8 MHz: 69.4 ± 7.8 mm). A decreased anechoic contrast was observed with a 4.5 cm depth for RIC5‐9‐D (7 MHz: 23.2 ± 1.8 dB, P > 0.05; 8 MHz: 17.7 ± 0.9 dB, P < 0.01) compared with IC5‐9‐D (7 MHz: 25.9 ± 1.2 dB; 8 MHz: 21.5 ± 0.8 dB). The contrast response and spatial resolution performance were comparable between the two transducers. RIC5‐9‐D showed comparable SNR of anechoic spheres compared to IC5‐9‐D.Conclusions2D images from a 3D probe exhibited comparable overall image quality for routine clinical pelvic imaging.

AbstractBecause of the rapidly growing use of ultrasound shear wave elastography (SWE) in clinical practices, there is a significant need for development of clinical physics performance assessment methods for this technology. This study aims to report two clinical medical physicists’ tasks: (a) acceptance testing (AT) of SWE function on ten commercial ultrasound systems for clinical liver application and (b) comparison of SWE measurements of targets across vendors for clinical musculoskeletal application. For AT, ten GE LOGIQ E9 XDclear 2.0 scanners with ten C1‐6‐D and ten 9L‐D transducers were studied using two commercial homogenous phantoms. Five measurements were acquired at two depths for each scanner/transducer pair by two operators. Additional tests were performed to access effects of different coupling media, phantom locations and operators. System deviations were less than 5% of group mean or three times standard deviation; therefore, all systems passed AT. A test protocol was provided based on results that no statistically significant difference was observed between using ultrasound gel and salt water for coupling, among different phantom locations, and that interoperator and intraoperator coefficient of variation was less than 3%. For SWE target measurements, two systems were compared — a Supersonic Aixplorer scanner with a SL10‐2 and a SL15‐4 transducer, and an abovementioned GE scanner with 9L‐D transducer. Two stepped cylinders with diameters of 4.05–10.40 mm were measured both longitudinally and transaxially. Target shear wave speed quantification was performed using an in‐house MATLAB program. Using the target shear wave speed deduced from phantom specs as a reference, SL15‐4 performed the best at the measured depth. However, it was challenging to reliably measure a 4.05 mm target for either system. The reported test methods and results could provide important information when dealing with SWE‐related tasks in the clinical environment.

AbstractWe evaluated a commercially available software package that uses B‐mode images to semi‐automatically measure quantitative metrics of ultrasound image quality, such as contrast response, depth of penetration (DOP), and spatial resolution (lateral, axial, and elevational). Since measurement of elevational resolution is not a part of the software package, we achieved it by acquiring phantom images with transducers tilted at 45 degrees relative to the phantom. Each measurement was assessed in terms of measurement stability, sensitivity, repeatability, and semi‐automated measurement success rate. All assessments were performed on a GE Logiq E9 ultrasound system with linear (9L or 11L), curved (C1‐5), and sector (S1‐5) transducers, using a CIRS model 040GSE phantom. In stability tests, the measurements of contrast, DOP, and spatial resolution remained within a ±10% variation threshold in 90%, 100%, and 69% of cases, respectively. In sensitivity tests, contrast, DOP, and spatial resolution measurements followed the expected behavior in 100%, 100%, and 72% of cases, respectively. In repeatability testing, intra‐ and inter‐individual coefficients of variations were equal to or less than 3.2%, 1.3%, and 4.4% for contrast, DOP, and spatial resolution (lateral and axial), respectively. The coefficients of variation corresponding to the elevational resolution test were all within 9.5%. Overall, in our assessment, the evaluated package performed well for objective and quantitative assessment of the above‐mentioned image qualities under well‐controlled acquisition conditions. We are finding it to be useful for various clinical ultrasound applications including performance comparison between scanners from different vendors.

Our goal in this work was to compare the results of common phantom tests made using matched and mixed ultrasound (US) scanner‐transducer combinations. Sets of common US quality assurance (QA) measurements were made using matched US scanner‐transducer combinations (i.e., transducers purchased for use with a particular scanner), as well as unmatched (mixed) combinations. Measurements of vertical and horizontal distance accuracy, and depth of penetration were performed using three common transducer types. Means, standard deviations, and differences between the mean mix and match measurements divided by the standard deviation (match‐mix difference, or MMD), and two‐sided, paired t‐tests were computed for the groups of mixed and matched measurements. MMDs for vertical and horizontal distance accuracy test results were less than 0.87 in all cases, well below our threshold value of 2.0, which indicates that a significant difference exists. MMDs for the depth of penetration measurements were less than 1.50, again below the threshold value. These results suggest that all of the mixed and matched data sets were very similar. The more sensitive t‐tests indicate statistically significant differences in only 2 of the 18 pairs of data sets. In conclusion, this study suggests that QA measurements generated by mixed or matched scanner‐transducer combinations are very comparable. The ability to obtain QA phantom test data from mixed scanner‐transducer combinations reduces the time required for US QA testing.PACS number(s): 87.57.–s, 87.62.+n