Single Ultrasound Image Research Articles

Validity of Using Still Images for Written Competency Assessments for Point-of-Care Ultrasound among Non-Radiologists-pilot study.

Background and Objective: Point-of-care ultrasound (POCUS) has become an integral part of the curricula of various spe-cialty residency programs. POCUS education requires learners to be proficient in image acqui-sition, image interpretation, and integration into clinical practice. The assessment of POCUS skills and competency is variable. This study aimed to assess the validity and reliability of us-ing still ultrasound images as examination test items without providing clinical vignettes or anatomical locations. Methods: This is a proof-of-concept survey pilot study, residents from the Middle East who had com-pleted their training in emergency medicine were surveyed during their competency assess-ment examination. They were provided with a single ultrasound image of an abdominal aortic aneurysm in both print and PowerPoint slide format. The task was to identify the diagnosis based on this image, which was taken from a reference emergency medicine textbook. Results: The study showed that only 50% of the residents correctly identified the diagnoses based on the still images. Many of the incorrect answers included diagnoses that were not related to the anatomical structure that was scanned, indicating that the residents were not able to identify the anatomical locations, which highlighted the challenges of interpreting a single still image without clinical context and emphasized the need for valid and reliable test items in point-of-care ultrasound (POCUS) education and assessment. Conclusions: National and international emergency medicine societies should establish standards for writ-ing POCUS examination test items to ensure their validity and reliability.

Reinforced Computer-Aided Framework for Diagnosing Thyroid Cancer.

Thyroid cancer is the most pervasive disease in the endocrine system and is getting extensive attention. The most prevalent method for an early check is ultrasound examination. Traditional research mainly concentrates on promoting the performance of processing a single ultrasound image using deep learning. However, the complex situation of patients and nodules often makes the model dissatisfactory in terms of accuracy and generalization. Imitating the diagnosis process in reality, a practical diagnosis-oriented computer-aided diagnosis (CAD) framework towards thyroid nodules is proposed, using collaborative deep learning and reinforcement learning. Under the framework, the deep learning model is trained collaboratively with multiparty data; afterward classification results are fused by a reinforcement learning agent to decide the final diagnosis result. Within the architecture, multiparty collaborative learning with privacy-preserving on large-scale medical data brings robustness and generalization, and diagnostic information is modeled as a Markov decision process (MDP) to get final precise diagnosis results. Moreover, the framework is scalable and capable of containing more diagnostic information and multiple sources to pursue a precise diagnosis. A practical dataset of two thousand thyroid ultrasound images is collected and labeled for collaborative training on classification tasks. The simulated experiments have shown the advancement of the framework in promising performance.

Validity of Using Still Images for Written Competency Assessments for Point-of-Care Ultrasound among Non-Radiologists-pilot study.

Background and Objective: Point-of-care ultrasound (POCUS) has become an integral part of the curricula of various spe-cialty residency programs. POCUS education requires learners to be proficient in image acqui-sition, image interpretation, and integration into clinical practice. The assessment of POCUS skills and competency is variable. This study aimed to assess the validity and reliability of us-ing still ultrasound images as examination test items without providing clinical vignettes or anatomical locations. Methods: This is a proof-of-concept survey pilot study, residents from the Middle East who had com-pleted their training in emergency medicine were surveyed during their competency assess-ment examination. They were provided with a single ultrasound image of an abdominal aortic aneurysm in both print and PowerPoint slide format. The task was to identify the diagnosis based on this image, which was taken from a reference emergency medicine textbook. Results: The study showed that only 50% of the residents correctly identified the diagnoses based on the still images. Many of the incorrect answers included diagnoses that were not related to the anatomical structure that was scanned, indicating that the residents were not able to identify the anatomical locations, which highlighted the challenges of interpreting a single still image without clinical context and emphasized the need for valid and reliable test items in point-of-care ultrasound (POCUS) education and assessment. Conclusions: National and international emergency medicine societies should establish standards for writ-ing POCUS examination test items to ensure their validity and reliability.

Classification of tumor in one single ultrasound image via a novel multi-view learning strategy

Computer-aided diagnosis (CAD) technology has been widely used in the early diagnosis of breast cancer. Nowadays, most of the existing breast ultrasound classification methods need to crop a tumor-centered image (TCI) on each image as the input of the system. These methods ignore the fact that the tumor as well as its surrounding tissues can actually be viewed from multiple aspects, and it is difficult to extract multi-resolution information applying only a single view image. In addition, the current methods do not effectively extract fine-grained features, and subtle details play an important role in breast classification. In our research, we propose a novel strategy to generate multi-resolution TCIs in a single ultrasound image, resulting in a multi-data-input learning task. Hence, a conventional single image based learning task is converted into a multi-view learning task, and an improved combined style fusion method suitable for a deep network is proposed, which integrates the advantage of the decision-based and feature-based methods to fuse the information of different views. At the same time, we first attempt to introduce the fine-grained classification method into breast classifications and capture the pairwise correlation between feature channels at each position to extract subtle information. The comparative experimental results show that our method can effectively improve the classification performance and achieves the best results in five metrics.

Reliability and validity of quantitative ultrasound for evaluating patellar alignment: A pilot study

BackgroundPatellar malalignment is a risk factor of patellofemoral pain. Evaluation of the patellar alignment have mostly used magnetic resonance imaging (MRI). Ultrasound (US) is a non-invasive instrument that can quickly evaluate patellar alignment. However, the method for evaluating patellar alignment via US has not been established. This study aimed to investigate the reliability and validity of evaluating patellar alignment via US. MethodsThe sixteen right knees were imaged via US and MRI. US images were obtained at two sites of the knee to measure US-tilt as the index of patellar tilt. Using a single US image, we measured US-lateral distance and US-angle as the index of patellar shift. All US images were obtained three times each by two observers to evaluate reliabilities. Lateral patellar angle (LPA), as the indicators of patellar tilt, and lateral patella distance (LPD) and bisect offset (BO), as the indicators of patellar shift, were measured via MRI. ResultsUS measurements provided high intra- (within-day and between days) and interobserver reliabilities with exception of interobserver reliability of US-lateral distance. Pearson correlation coefficient indicated that US-tilt is significantly positively correlated with LPA (r = 0.79), and US-angle is significantly positively correlated with LPD (r = 0.71) and BO (r = 0.63). ConclusionEvaluating patellar alignment via US showed high reliabilities. US-tilt and US-angle showed moderate to strong correlation with MRI indices of patellar tilt and shift via MRI, respectively. US methods are useful for evaluating accurate and objective indices of patellar alignment.

Ultrasonographic pathological grading of prostate cancer using automatic region-based Gleason grading network.

The Gleason scoring system is a reliable method for quantifying the aggressiveness of prostate cancer, which provides an important reference value for clinical assessment on therapeutic strategies. However, to the best of our knowledge, no study has been done on the pathological grading of prostate cancer from single ultrasound images. In this work, a novel Automatic Region-based Gleason Grading (ARGG) network for prostate cancer based on deep learning is proposed. ARGG consists of two stages: (1) a region labeling object detection (RLOD) network is designed to label the prostate cancer lesion region; (2) a Gleason grading network (GNet) is proposed for pathological grading of prostate ultrasound images. In RLOD, a new feature fusion structure Skip-connected Feature Pyramid Network (CFPN) is proposed as an auxiliary branch for extracting features and enhancing the fusion of high-level features and low-level features, which helps to detect the small lesion and extract the image detail information. In GNet, we designed a synchronized pulse enhancement module (SPEM) based on pulse-coupled neural networks for enhancing the results of RLOD detection and used as training samples, and then fed the enhanced results and the original ones into the channel attention classification network (CACN), which introduces an attention mechanism to benefit the prediction of cancer grading. Experimental performance on the dataset of prostate ultrasound images collected from hospitals shows that the proposed Gleason grading model outperforms the manual diagnosis by physicians with a precision of 0.830. In addition, we have evaluated the lesions detection performance of RLOD, which achieves a mean Dice metric of 0.815.

Reliability and Comparison of Sonographic Methods for In Vivo Measurement of Human Biceps Femoris Long-Head Architecture.

Biceps femoris long-head fascicle length ( Lf ), fascicle angle (FA), and muscle thickness (MT) estimates obtained across 2 d from extended field-of-view (EFOV) sonographic images were compared with those measured from a collage of single ultrasound images (to visualize entire fascicles) as well as a range of geometric equations and extrapolation methods used on single images. Both test validity and intraday reliability were determined. Twenty healthy adults (10 men and 10 women) were tested on two occasions (day 1 and day 2), 7 d apart at the same time of day for test-retest measurements. Ultrasound imaging was performed using EFOV and static image acquisition sequences; in the latter, four single images were acquired in-series along the muscle. From these images, Lf was assessed using seven methods: EFOV, collage, manual linear extrapolation, and four different trigonometric equations (termed equations A, B, C, and D), and FA and MT were measured in EFOV, collage, and single images. Lf , FA, and MT measured on days 1 and 2 were not different ( P > 0.05) for any method, reliabilities were very high (intraclass correlation coefficient, 0.91-0.98), and correlations were strong (≥0.84). Significant correlations ( P < 0.05; r = 0.67-0.98) were found between EFOV and the other measurement techniques for Lf , FA, and MT. The collage method had the highest reliability for Lf , and highest rank order and correlation with EFOV. Although the six different techniques used to estimate Lf provided values similar to EFOV, higher between-subject measurement variability was observed with trigonometric equations, and the collage method described herein provided the most accurate and reliable results and is therefore recommended for biceps femoris long-head architectural analysis when EFOV is not available.

Predicting Breast Tumor Malignancy Using Deep ConvNeXt Radiomics and Quality-Based Score Pooling in Ultrasound Sequences.

Breast cancer needs to be detected early to reduce mortality rate. Ultrasound imaging (US) could significantly enhance diagnosing cases with dense breasts. Most of the existing computer-aided diagnosis (CAD) systems employ a single ultrasound image for the breast tumor to extract features to classify it as benign or malignant. However, the accuracy of such CAD system is limited due to the large tumor size and shape variation, irregular and ambiguous tumor boundaries, and low signal-to-noise ratio in ultrasound images due to their noisy nature and the significant similarity between normal and abnormal tissues. To handle these issues, we propose a deep-learning-based radiomics method based on breast US sequences in this paper. The proposed approach involves three main components: radiomic features extraction based on a deep learning network, so-called ConvNeXt, a malignancy score pooling mechanism, and visual interpretations. Specifically, we employ the ConvNeXt network, a deep convolutional neural network (CNN) trained using the vision transformer style. We also propose an efficient pooling mechanism to fuse the malignancy scores of each breast US sequence frame based on image-quality statistics. The ablation study and experimental results demonstrate that our method achieves competitive results compared to other CNN-based methods.

Multioperator Algorithm-Based Ultrasound Imaging Combined with Hysteroscopy for the Diagnosis of Endometrial Polyps

This study aimed to explore the diagnosis of endometrial polyps (EMP) by ultrasound imaging based on multi-operator algorithms combined with hysteroscopy. This study is the first to investigate the resolution of the ultrasound adaptive beamforming algorithm (MOAD), after which the proposed algorithm was applied to ultrasound diagnosis of 102 patients with EMP and pathologically diagnosed with vaginal irregular bleeding, and the evaluation efficacy of the MOAD algorithm based on EMP was compared. The resolution of the MOAD-based imaging algorithm (0.0645) was significantly lower than that of the diagonal loading operator (0.1475), the symbol coherence coefficient operator (0.1342), and the generalized coherence factor operator (0.1234), with significant differences ( P &lt; 0.05 ). The proportion of patients with EMP aged 46–55 years was the largest (55.9%). There were 64 cases of EMP that produced complications, of which the proportion of patients with uterine fibroids (41.52%), abnormal uterine bleeding (76.24%), and menstrual changes (42.57%) was relatively large. Patients with nonfunctioning polyps accounted for the largest proportion (84.46%), followed by those with basal polyps (76.24%), and the difference was statistically significant ( P &lt; 0.05 ). The positive cases of EMP detected by ultrasound imaging (38 cases) were significantly lower than those with pathological diagnosis (94 cases), and the difference was statistically significant ( P &lt; 0.05 ). The SE, SP, FNR, and FPR of EMP diagnosed by ultrasound imaging combined with hysteroscopy were 64.45%, 84.67%, 35.48%, and 13.36%, respectively. It has high diagnostic value compared with single ultrasound imaging diagnosis, and the difference was statistically significant ( P &lt; 0.05 ). In conclusion, the imaging based on the MOAD algorithm is obvious and the pixel resolution can be successfully improved. The diagnostic value of ultrasound combined with hysteroscopy for EMP was better than that of ultrasound alone ( P &lt; 0.05 ), and it had a high diagnostic value.

Changes in Biceps Femoris Long Head Fascicle Length after 10-d Bed Rest Assessed with Different Ultrasound Methods.

This study aimed to investigate the changes in fascicle length (Lf) of biceps femoris long head (BFlh) after 10 d of bed rest (BR) by comparing four different ultrasound (US) methods. Ten healthy men participated in 10-d BR. Before (BR0) and after (BR10) the BR period, BFlh Lf values were obtained using 1) extended-field-of-view (EFOV) technique, 2) the manual linear extrapolation (MLE) method, and 3) two trigonometric equations (equations A and B) from a single US image. After BR10, decreased Lf values were observed by EFOV (P < 0.001; Hedges' g = 0.29) and MLE (P = 0.0082; g = 0.22) methods, but not with equations A and B. Differences between equation A and the other US methods were detected at both time points. The percentage of changes in Lf between BR0 and BR10 was influenced by the US methods applied, with difference detected between the changes measured by EFOV and the ones estimated by equation A (P = 0.04; g = 0.53). Bland-Altman analyses revealed relevant average absolute biases in Lf between EFOV and other methods at both time points (range BR0-BR10: MLE, 0.3-0.37 cm (3.4%-4.32%); equation B, 0.3-0.48 cm (3.24%-5.41%); equation A, 2.44-2.97 cm (24.05%-29.2%)). A significant correlation (r = 0.83) in percentage of change in Lf values was observed only between EFOV and MLE. We showed that four distinct US methods lead to different results in the assessment of BFlh Lf changes after a short-term period of unloading. The implementation of EFOV technique (or alternatively MLE) to assess Lf changes in BFlh during longitudinal studies is warranted.

Efficient automatic 2D/3D registration of cardiac ultrasound and CT images

ABSTRACT Hybrid ablations are a promising but difficult intervention for the treatment of atrial fibrillation. With the ultimate goal of providing navigation support for such procedures, we investigate a registration algorithm for routinely available preoperative and intraoperative images. We propose a fully automatic segmentation algorithm for the boundaries of cardiac chambers in intraoperative TEE ultrasound using a generic heart model. The resulting ultrasound segmentations are initially registered to the preoperative CT model using a frame-to-slice search, which is then refined using an efficient continuous optimisation. Results are presented for data sets from three patients who underwent hybrid ablations at our institution. The mean time to process a single ultrasound image from segmentation to registration with CT was 1.5 s, with the mean RMS error across the sequence being 4.8 mm. With further validation, these results show promise for surgical navigation in hybrid ablation procedures.

Integrating Innovative Handheld Portable Cordless Ultrasound‐Probe‐System with Artificial Intelligence for Pre‐Healthcare Students in Human Cadaver Lab Course

INTRODUCTIONThe importance of ultrasound regarding its depth and breadth across medical fields continues to expand in development and use while affordability improves for the user and patient. Sonoanatomy is important for interpreting the anatomical architecture adjacent to the probe. Medical educators responsible for teaching ultrasound are faced with two obstacles, teaching how to drive the ultrasound technology and understanding the surface, static and dynamic sonoanatomy of the regions of the body.AIMThe aim of this study was to identify and teach sonoanatomy with ultrasound which limits or removes driving the knobology associated with traditional ultrasound technology.METHODSLiterature search was conducted to identify an ultrasound probe system which is handheld, cordless, with no buttons and uses artificial intelligence to provide diagnostic quality imaging. An upper division baccalaureate semester cadaver dissection course (pre healthcare Gen Z students N=14) which included sonoanatomy teaching and examination with a traditional cart ultrasound probe system was the arena to introduce such technology. Students rotated during every dissection lab to practice ultrasound probe positioning, image acquisition and anatomy interpretation.RESULTSClarius handheld wireless multiprobe ultrasound‐probe‐system compatible with Apple and Android mobile devices was identified and used equally with traditional cart ultrasound. Students were successfully taught surface and regional sonoanatomy in lectures and tutorials using exam results.DISCUSSIONClarius is shock and water resistant and utilizes automated AI with a real time automated time‐gain‐compensation (TGC) engine. Traditional ultrasound systems require users adjusting several TGC controls (knobs) to optimize imaging and often when changing scanning planes, attenuation variance forces users to adjust TGC often resulting in failed images adding to the learning curve. Clarius automated AI overrides user interaction with TGC controls if desired. Analyzing each single ultrasound image captured during scanning at rates of 30 frames per second, gain can be tuned with high level precision to 1mm, all in real‐time, with instant feedback resulting in quality images.CONCLUSIONStudents preferred not having to adjust knobs and performed better acquiring quality images with Clarius probe system on cadaver and simulation exercises then with a traditional ultrasound cart system. This pilot project suggests more Clarius ultrasound studies could be used to train novices in healthcare.

Automatic diagnosis for thyroid nodules in ultrasound images by deep neural networks.

Thyroid cancer is a disease in which the first symptom is a nodule in the thyroid region of the neck. It is one of the cancers with the highest incidences, and has the highest increase rate in the last thirty years. Ultrasonography is one of the most sensitive and widely used methods for detecting thyroid nodules. To assist in the analysis of thyroid ultrasound images, many computer-aided diagnosis methods have been proposed. Most of these methods perform diagnosis using only a single ultrasound image instead of using all images from an examination, which loses the overall information related to the thyroid nodules. However, in an ultrasound examination, the sonographer analyzes the thyroid nodule based on multiple images from different views. In the current study, a deep learning method is proposed to diagnose thyroid nodules using multiple ultrasound images in an examination as input. An attention-based feature aggregation network is proposed to automatically integrate the features extracted from multiple images in one examination, utilizing different views of the nodules to improve the performance of recognizing malignant nodules in the ultrasound images. To train and evaluate the proposed method, a large dataset is constructed. The experimental results demonstrate that our method achieves comparable performance with state-of-the-art methods for the diagnosis of thyroid ultrasound images.

Ultrasound-derived Biceps Femoris Long Head Fascicle Length: Extrapolation Pitfalls.

This study aimed to compare biceps femoris long head (BFlh) fascicle length (Lf) obtained with different ultrasound-based approaches: 1) single ultrasound images and linear Lf extrapolation, 2) single ultrasound images and one of two different trigonometric equations (termed equations A and B), and 3) extended field of view (EFOV) ultrasound images. Thirty-seven elite alpine skiers (21.7 ± 2.8 yr) without a previous history of hamstring strain injury were tested. Single ultrasound images were collected with a 5-cm linear transducer from BFlh at 50% femur length and were compared with whole muscle scans acquired by EFOV ultrasound. The intrasession reliability (intraclass correlation coefficient [ICC3,k]) of Lf measurements was very high for both single ultrasound images (i.e., Lf estimated by linear extrapolation; ICC3,k = 0.96-0.99, SEM = 0.18 cm) and EFOV scans (ICC3,k = 0.91-0.98, SEM = 0.19 cm). Although extrapolation methods showed cases of Lf overestimation and underestimation when compared with EFOV scans, mean Lf measured from EFOV scans (8.07 ± 1.36 cm) was significantly shorter than Lf estimated by trigonometric equations A (9.98 ± 2.12 cm, P < 0.01) and B (8.57 ± 1.59 cm, P = 0.03), but not significantly different from Lf estimated with manual linear extrapolation (8.40 ± 1.68 cm, P = 0.13). Bland-Altman analyses revealed mean differences in Lf obtained from EFOV scans and those estimated from equation A, equation B, and manual linear extrapolation of 1.91 ± 2.1, 0.50 ± 1.0, and 0.33 ± 1.0 cm, respectively. The typical extrapolation methods used for estimating Lf from single ultrasound images are reliable within the same session, but not accurate for estimating BFlh Lf at rest with a 5-cm field of view. We recommend that EFOV scans are implemented to accurately determine intervention-related Lf changes in BFlh.

Bilateral testicular micro lithiasis in pediatrics secondary to testicular trauma

Introduction: Testicular micro lithiasis is defined as the presence of five or more testicular microcalcifications in a single ultrasound image. The clinical meaning is still unknown. It has been associated with: cryptorchidism, testicular torsion, varicocele, gonadal dysgenesis or chromosomal disorders such as Klinefelter syndrome, Down syndrome, McCune-Albright syndrome and Peuzt-Jeghers syndrome, which may be associated with testicular deterioration and infertility. Clinical case: A male patient of 10years. I present testicular trauma at the age of 5years 2months. With peri testicular hematoma, peri testicular and scrotal hematoma drainage was performed without incident. Its evolution was adequate and it was withdrawn at 48 hours after surgery. Five years after the surgery, he refers to inguinoscrotal pain. Physical examination revealed: testicular atrophy and mild hydrocele. Bilateral testicular ultrasound was performed. With the following findings: testicular micro-lithiasis in the parenchyma of both testicles. Discussion: In 2010, a meta-analysis did not report a causal link between testicular micro lithiasis and cancer in the absence of additional risk factors. Conclusion: There is not enough evidence in the current literature to support any clinical surveillance regimen in patients with testicular micro lithiasis. Self-examination is the most important factor in the early detection of testicular malignancy. Keywords: testicular micro lithiasis, testicular cáncer, microlithiasis

Prediction and Validation of DXA-Derived Appendicular Fat-Free Adipose Tissue by a Single Ultrasound Image of the Forearm in Japanese Older Adults.

To develop regression-based equations for estimating dual-energy x-ray absorptiometry (DXA) derived appendicular fat-free adipose tissue (FFAT) using a single ultrasound image in the forearm, and to investigate the validity of those equations to calculate FFAT-free appendicular lean mass (aLM-minus-FFATappendicular ) in 311 Japanese adults aged 60 to 79 years. Subjects were randomly separated into two groups: 215 in the model-development group (91 men and 124 women) and 96 in the cross-validation group (42 men and 54 women). Appendicular fat mass and aLM were measured by the DXA, and subcutaneous adipose tissue (AT-forearm) and muscle (MT-ulna) thicknesses were measured by ultrasound. Appendicular FFAT was calculated based on the results of a previous study (appendicular FFAT = appendicular fat mass/0.85 x 0.15). The aLM was estimated from MT-ulna using a previously published equation (aLM = 4.89 x MT-ulna x body height - 9.15). Stepwise linear regression analysis was used to determine predictive models for DXA-derived appendicular FFAT from AT-forearm, sex, age, and anthropometrical variables. The best ultrasound prediction equation for estimation of appendicular FFAT was developed and then cross-validated in a subsample of older adults. There was no significant difference between the DXA-derived and ultrasound-predicted aLM-minus-FFATappendicular . A strong correlation was observed between the DXA-derived and ultrasound-predicted aLM-minus-FFATappendicular (r = 0.935, P < .001). Bland-Altman analysis did not indicate a bias in the prediction of the aLM-minus-FFATappendicular for the validation group. Our results indicated that a single ultrasound forearm measurement can be used to accurately estimate DXA-derived aLM-minus-FFATappendicular in Japanese older adults, which may be advantageous for community-based physical examinations.

Criterion Validity of Ultrasound Imaging: Assessment of Multifidi Cross-Sectional Area in Older Adults With and Without Chronic Low Back Pain

Ultrasound (US) imaging may be a cost-conscious alternative to magnetic resonance imaging (MRI), which is the criterion standard for muscle cross-sectional area (CSA) assessment. Within the trunk, when compared with MRI, US has been shown to be valid for assessing lumbar multifidi CSA in younger, asymptomatic individuals. To date, there are no studies validating US for multifidi CSA assessment in older adults or individuals with low back pain. Given age- and pain-related muscle changes, validation of US is needed in these populations. If valid for multifidi CSA assessment, US may be used to evaluate short-term changes in muscle size in response to exercise-based interventions among older adults. The primary objective of this study was to evaluate the validity of US for multifidi CSA assessment as compared with MRI in older adults with and without chronic low back pain (CLBP). The secondary objective was to determine whether a single US image was valid for assessment of multifidi CSA or whether the average of 3 US images should be recommended. Twenty community-dwelling older adults (ie, 10 with and 10 without CLBP), ages 60 to 85 years, were recruited. US images and MRI slices of multifidi muscle were obtained and L4 multifidi CSAs were measured. Intraclass correlation coefficients (ICCs) were calculated to assess agreement between MRI measures and a single US image and MRI measures and the average of 3 US images. ICC point estimates were excellent for older adults with CLBP for a single US image (ICCs = 0.90-0.97), but ICC point estimates for participants without CLBP ranged from fair to excellent (ICCs = 0.48-0.86). ICC point estimates for the average of 3 US images for both groups were better than for a single image (ICCs = 0.95-0.99). For assessment of L4 multifidi CSA, US is a valid alternative to MRI for older adults with and without CLBP. However, limitations of US, such as the inability to quantify intramuscular fat, which may be increased with aging and CLBP, should be considered. CSA measurement of 3 US images, rather than a single image, is recommended.

Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study

Measurement reliability of the L4/5 lumbar multifidus (MF) muscles is high in older adults, but few studies have investigated measurement reliability of the abdominal and upper lumbar MF muscles in this age group. ObjectivesTo determine measurement reliability of abdominal and lumbar MF muscles from a single ultrasound (US) image in older adults. MethodsResting thickness of rectus abdominis and obliquus externus, resting and contracted thickness of obliquus internus, transversus abdominis and lumbar MF, and resting cross-sectional area (CSA) of MF levels (L2-5) were obtained from US images of 92 community-dwelling older adults (aged 65–89 years). Measurements of images were undertaken by an experienced rater and repeated 7–10 days later for intra-rater, and by a second expert rater for inter-rater calculations. Intra-rater reliability was estimated for all muscles. Inter-rater reliability was estimated for all abdominal muscles and for L5 multifidus. Reliability was estimated by intraclass correlation coefficients (ICC). ResultsIntra-rater ICC(3,1) and inter-rater ICC(2,1) of resting thickness measures of all muscles and CSA of MF were ≥0.86. The ICCs for percentage thickness change were ≥0.76 for the abdominal muscles, and ≥0.42 for MF. ConclusionsMeasurement reliability of US imaging for abdominal and MF muscle thickness and MF CSA was high, and consistent with previous findings for younger adults. Reliability of percentage thickness change was lower suggesting caution is needed when using this as an outcome measure or study factor among older adults.

Grey-Scale Analysis Improves the Ultrasonographic Evaluation of Thyroid Nodules.

Ultrasonography is the main imaging method for the workup of thyroid nodules. However, interobserver agreement reported for echogenicity and echotexture is quite low. The aim of this study was to perform quantitative measurements of the degree of echogenicity and heterogeneity of thyroid nodules, to develop an objective and reproducible method to stratify these features to predict malignancy.A retrospective study of patients undergoing ultrasonography-guided fine-needle aspiration was performed in an University hospital thyroid center. From January 2010 to October 2012, 839 consecutive patients (908 nodules) underwent US-guided fine-needle aspiration. In a single ultrasound image, 3 regions of interest (ROIs) were drawn: the first including the nodule; the second including a portion of the adjacent thyroid parenchyma; the third, the strap muscle. Histogram analysis was performed, expressing the median, mean, and SD of the gray levels of the pixels comprising each region. Echogenicity was expressed as a ratio: the nodule/parenchyma, the nodule/muscle, and parenchyma/muscle median gray ratios were calculated. The heterogeneity index (HI) was calculated as the coefficient of variation of gray histogram for each of the 3 ROIs. Cytology and histology reports were recorded.Nodule/parenchyma median gray ratio was significantly lower (more hypoechoic) in nodules found to be malignant (0.45 vs 0.61; P = 0.002) and can be used as a continuous measure of hypoechogenicity (odds ratio [OR] 0.12; 95% confidence interval [CI] 0.03–0.49). Using a cutoff derived from ROC curve analysis (<0.46), it showed a substantial inter-rater agreement (k = 0.74), sensitivity of 56.7% (95% CI 37.4–74.5%), specificity of 72.0% (67.8–75.9%), positive likelihood ratio (LR) of 2.023 (1.434–2.852), and negative LR of 0.602 (0.398–0.910) in predicting malignancy (diagnostic odds ratio 3.36; 1.59–7.10). Parenchymal HI was associated with anti-thyroperoxidase positivity (OR 19.69; 3.69–105.23). The nodule HI was significantly higher in malignant nodules (0.73 vs 0.63; P = 0.03) and, if above the 0.60 cutoff, showed sensitivity of 76.7% (57.7–90.1%), specificity of 46.8% (42.3–51.4%), positive LR of 1.442 (1.164–1.786), and negative LR of 0.498 (0.259–0.960).Evaluation of nodule echogenicity and echotexture according to a numerical estimate (nodule/parenchyma median gray ratio and nodule HI) allows for an objective stratification of nodule echogenicity and internal structure.

A coarse-to-fine IP-driven registration for pose estimation from single ultrasound image

A fast registration making use of implicit polynomial (IP) models is helpful for the real-time pose estimation from single clinical free-hand Ultrasound (US) image, because it is superior in the areas such as robustness against image noise, fast registration without enquiring correspondences, and fast IP coefficient transformation. However it might lead to the lack of accuracy or failure registration.In this paper, we present a novel registration method based on a coarse-to-fine IP representation. The approach starts from a high-speed and reliable registration with a coarse (of low degree) IP model and stops when the desired accuracy is achieved by a fine (of high degree) IP model. Over the previous IP-to-point based methods our contributions are: (i) keeping the efficiency without requiring pair-wised correspondences, (ii) enhancing the robustness, and (iii) improving the accuracy. The experimental result demonstrates the good performance of our registration method and its capabilities of overcoming the limitations of unconstrained freehand ultrasound data, resulting in fast, robust and accurate registration.