3D US Research Articles

Automated diagnosis of hip dysplasia from 3D ultrasound using artificial intelligence: A two-center multi-year study

BackgroundThree-dimensional (3D) Ultrasonography of the hip introduces opportunities for automation by artificial intelligence (AI) to diagnose developmental dysplasia of the hip (DDH). We compared conventional clinical diagnosis and a commercial FDA-cleared AI algorithm in detecting and classifying hip dysplasia on 3DUS. PurposeTo validate the accuracy of AI in detection and classification of DDH from 3D US. Materials and methodsInfant hip 3D US obtained prospectively for clinical suspicion of DDH at two tertiary pediatric hospitals (in Edmonton, Canada and Melbourne, Australia) from January 2012 to December 2020 were analyzed retrospectively by an AI algorithm trained on other data from Edmonton and not previously exposed to Melbourne images. Results were compared to the clinical diagnosis made using concurrently obtained conventional two-dimensional (2D) US of the same hips. The AI algorithm automatically calculated two indices (alpha angle and femoral head coverage) without user input and classified hips as normal or dysplastic. Diagnostic accuracy of AI for human expert reference-standard diagnosis, and inter-rater reliability between AI and clinical diagnosis were assessed. Results2492 hips from 1563 patients were evaluated (1294 hips in Edmonton, 1198 in Melbourne; 1061 (68%) female patients, mean age 87 days, range 4–267 days). 2327/2492 hips were clinically classified as Normal or Borderline (initially Graf IIa but spontaneously normalizing at follow-up imaging), while 165 hips were clinically Dysplastic and treated. AI correctly categorized 90% (148/165) of Dysplastic hips and 86% (2006/2327) of the other hips overall, with sensitivity 0.90 for dysplasia. Only one case of severe dysplasia (Graf III) was missed by AI. Agreement between AI and clinical diagnosis was very high (kappa = 0.79). ConclusionAutomated artificial intelligence analysis of 3D ultrasound images correlated well to human expert clinical diagnosis of hip dysplasia from concurrent conventional 2D ultrasound of the same hips.

Population-based 3D respiratory motion modelling from convolutional autoencoders for 2D ultrasound-guided radiotherapy.

Radiotherapy is a widely used treatment modality for various types of cancers. A challenge for precise delivery of radiation to the treatment site is the management of internal motion caused by the patient's breathing, especially around abdominal organs such as the liver. Current image-guided radiation therapy (IGRT) solutions rely on ionising imaging modalities such as X-ray or CBCT, which do not allow real-time target tracking. Ultrasound imaging (US) on the other hand is relatively inexpensive, portable and non-ionising. Although 2D US can be acquired at a sufficient temporal frequency, it doesn't allow for target tracking in multiple planes, while 3D US acquisitions are not adapted for real-time. In this work, a novel deep learning-based motion modelling framework is presented for ultrasound IGRT. Our solution includes an image similarity-based rigid alignment module combined with a deep deformable motion model. Leveraging the representational capabilities of convolutional autoencoders, our deformable motion model associates complex 3D deformations with 2D surrogate US images through a common learned low dimensional representation. The model is trained on a variety of deformations and anatomies which enables it to generate the 3D motion experienced by the liver of a previously unseen subject. During inference, our framework only requires two pre-treatment 3D volumes of the liver at extreme breathing phases and a live 2D surrogate image representing the current state of the organ. In this study, the presented model is evaluated on a 3D+t US data set of 20 volunteers based on image similarity as well as anatomical target tracking performance. We report results that surpass comparable methodologies in both metric categories with a mean tracking error of 3.5±2.4mm, demonstrating the potential of this technique for IGRT.

Effect of knee joint angle on individual hamstrings morphology quantified using free-hand 3D ultrasonography

Exercise responses and injury rates differ between individual hamstrings and this may be linked with their morphology. The aim of this study was to compare muscle length and tendon dimensions between the individual hamstrings at two knee joint angles using free hand three-dimensional ultrasound (3D US). Muscle-tendon length and distal tendon cross-sectional area (CSA), volume, length and echogenicity of biceps femoris long (BFlh) and short (BFsh) head, semimembranosus (SM) and semitendinosus (ST) of 16 individuals were measured using free-hand 3D US at 0° (full extension) and 45° of knee flexion. ST showed the greatest length than all muscles and BFsh the lowest (p < 0.05). No difference was observed between SM and BFlh length (p > 0.05). Of the four muscles, ST tendon was longer, with less volume and CSA but greater echogenicity than the other tendons. In contrast, SM and BFlh showed shorter tendons and lower echogenicity but a greater volume and CSA than ST (p < 0.05). Muscle and tendon lengthened from 45° to 0° knee flexion angle (p < 0.05) but this change was not statistically different between individual hamstrings (p > 0.05). Freehand 3D US indicated that hamstring muscle length and distal tendon dimensions differ between individual hamstrings. All muscles and tendons lengthened as the knee was extended but this change was similar for all individual hamstrings.

Prenatal prediction of respiratory distress syndrome by multimodality approach using 3D lung ultrasound, lung-to-liver intensity ratio tissue histogram and pulmonary artery Doppler assessment of fetal lung maturity

Studying the correlation of different lung parameters, using three-dimensional ultrasound (3D US) with fetal lung maturity (FLM) to predict the development of neonatal respiratory distress syndrome (RDS). Three-dimensional ultrasound was done to record the fetal lung volume (FLV), fetal lung-to-liver intensity ratio (FLLIR) and the main pulmonary artery (MPA) blood flow parameters; pulsatility index (PI), resistive index (RI) and acceleration time-to-ejection time ratio (At/Et), to 218 women between 32 and 40 weeks gestational age within 24 h from labor. Of 218 fetuses examined, final analysis was done for 143 fetuses. Thirty eight (26.5%) were diagnosed with RDS. The MPA PI and RI were significantly higher in fetuses diagnosed with RDS compared with those without (2.51 ± 0.33 and 0.90 ± 0.03 cm/s versus 1.96 ± 0.20 and 0.84 ± 0.01 cm/s; p value < 0.001 and <0.001 respectively). MPA At/Et was significantly lower (0.24 ± 0.04 vs 0.35 ± 0.04; p value < 0.001). FLLIR was significantly lower (1.04 ± 0.07 vs 1.18 ± 0.11; p value < 0.001), and the mean FLV was significantly smaller (28.23 ± 5.63, vs 38.87 ± 4.68 cm3; p value < 0.001). Main pulmonary artery (PI, RI, At/Et ratio), FLIIR, and mean FLV can be used as reliable predictors of neonatal RDS. 3D ultrasound VOCAL technique, ultrasound tissue histogram and pulmonary artery Doppler are reliable tools for prenatal prediction of fetal lung maturity.

Müllerian Duct Anomalies: Role in Fertility and Pregnancy.

Müllerian duct anomalies (MDAs) have important implications for the reproductive health of female patients. In patients with both infertility and recurrent pregnancy loss, the incidence of MDAs is as high as 25%. Congenital uterine anomalies are often only part of a complex set of congenital anomalies involving the cervix, vagina, and urinary tract. Multiple classification systems for MDAs exist, each with different criteria that vary most for the diagnosis of septate uterus. Recognizing the features that guide clinical management is essential for interpretation. Identification of an MDA should prompt evaluation for associated urinary tract anomalies. In patients with infertility who seek to use assisted reproductive technologies such as intrauterine insemination, recognition of MDAs may have an affect on reproductive success, particularly in patients who have an incomplete and clinically occult septum that divides the cervix. Two-dimensional US is the first-line modality for evaluating the uterus and adnexa. Three-dimensional (3D) US or MRI may help to visualize the external uterine fundal contour and internal indentation of the endometrial cavity, which are two morphologic characteristics that are keys to the diagnosis of congenital uterine anomalies. Hysterosalpingo contrast-enhanced US may be performed in conjunction with 3D US to evaluate uterine morphologic characteristics, the endometrial cavity, and tubal patency in a single examination. MRI helps to characterize rudimentary uteri in patients with müllerian hypoplasia and allows assessment for ectopic ureters, abnormally positioned ovaries, or associated deep infiltrative endometriosis. Online supplemental material is available for this article. ©RSNA, 2021.

VP15.04: Use of 3D during screening ultrasounds to improve diagnosis of cleft palate: feasibility and performance according to operator experience

Prenatal diagnosis of cleft palate is often missed by conventional 2D US, especially if not associated with cleft lip. This has implications as cleft palate can be associated with chromosomal abnormalities or syndromes. A higher detection rate (DR) would allow the performance of diagnostic tests, counselling, planning of follow-up. The aims of the study were to assess the rate of successful acquisition of 3D volumes of the fetal palate during screening-US and the quality of the volume obtained according to the operator's experience. Patients referred for anomaly and growth scans underwent study of the fetal palate with 2D and 3D only if feasible and within the allocated time of the examination with Samsung VS80 or GE Voluson E10 machines. Different render options were compared (Studio live HD/Surface/Skeleton for GE and Realistic view/CrystalVue for Samsung). Examinations were performed by 3 operators with different levels of expertise. The reasons for failed acquisition were recorded and volumes were assessed by 2 external operators and scored for quality (0-6 points). 188 patients were included (mean GA of 23,5 weeks+/-4,7). Volumes were acquired in 10-59% of cases, with the worst rate for the operator with < 5years of experience (p< 0.05). Lack of time was the most frequent reason for acquisition failure (58,2%), followed by malposition (24,5%) and BMI (17,3%). GA and BMI had an important role in the quality of the image obtained. The more experienced operators had better scores, with images considered satisfactory in nearly 80% of cases, compared to 60% of the less expert operator (n.s). Volume acquisition of the palate can be implemented in screening scans. This can be performed even by less-expert operators obtaining good quality volumes in more than half of the cases. However, despite the potential of this technique in increasing the DR of cleft palate, the frequent need for extra time, fetal malposition, and the need for high-quality equipment are obstacles to a formal recommendation.

O-025 Value of hysteroscopy

Abstract text The 3 main characters in any ART are the uterus, the endometrium, and the embryo. With the rutinary use of 2D US and especially 3D US in infertile patients we are able to assess most of the uterine anomalies and intrauterine pathologies. Until recently, the assessment of the endometrial cavity with hysteroscopy was reserved only for cases of IVF failure, as a complementary evaluation to ensure that nothing was missed during the ultrasound scan. We also need to remember how hysteroscopy was performed in the past and, unfortunately, in some cases even today. In the old days, to assess the endometrial cavity, it was required to take the patient to the operating room and under general anesthesia, dilate the cervix to then introduce a large diameter hysteroscope only for diagnostic purposes. The prevalent working model in ART today is in close collaboration between Fertility specialists, Ultrasound and Hysteroscopy units in order to improve patient’s outcome. In our center, a diagnostic hysteroscopy is performed as part of the diagnostic workup of the infertility patient. It is performed in office setting and without anesthesia, by the staff of the reproduction and infertility unit. When we look at hysteroscopy as an in-office procedure and no longer as a procedure performed in operating room, we appreciate that it is something that all infertility patients can benefit from. The right question to ask now would be “Is there a benefit in performing a hysteroscopy to all our patients?”. Ultrasound is not perfect, especially when evaluating the endometrium. The gold standard and the only modality that we have to assess the endometrium with direct visualization is hysteroscopy. Chronic endometritis, adhesions and adenomyosis are only a few examples of what can be assessed by direct vision of the endometrial cavity. The advantages of a simple, inexpensive, office procedure such as hysteroscopy outweigh any other consideration against it. During my talk all these points will be presented helping to understand why hysteroscopy is becoming an indispensable tool in every assisted fertility unit.

Comparative Study between the Role of Two dimensional and Three dimensional Ultrasound in Assessment of Fetal Central Nervous System Congenital Anomalies

ABSTRACT Objective: This study aimed to evaluate the role of three dimensional (3D) ultrasound compared with two dimensional (2D) ultrasound in diagnosis of fetal central nervous system congenital malformations. Patients and Methods: this Prospective study was performed on total 100 patients with gestational age from 18 weeks to 24 weeks who underwent 2D ultrasound then examined by 3D ultrasound at Obstetrics and Gynecology Department, Bab el-Sheria, Al-azhar University Hospital. Results: Statistical analysis revealed that the most common neurological malformations detected were Holoprosencephaly and Anencephaly which had significant high agreement between 2D and 3D regarding their diagnosis. There was significant agreement between 2D and 3D regarding the congenital fetal spinal and cranial malformations. The 2D and 3D ultrasound provided equal diagnostic information in microcephaly while 3D US provided more diagnostic information than 2D US for Choroid plexus cyst, agenesis of the corpus callosum and spina bifida. Our study revealed that the 3D US was superior to 2D US in identifying small cranial malformations as choroid plexus cyst and the spine e.g Spina bifida. Conclusion: In comparison with 2D US, 3D US improves the diagnostic capability by illustrating more diagnostic details in evaluation of malformations of the fetus, especially in visualization of fetal malformations of the small cranial anomalies and spina bifida . In addition to 2D US, 3D US is a valuable adjunct in prenatal detection of anomalies.

Assessment of the fetal thymus gland: Comparing MRI-acquired thymus volumes with 2D ultrasound measurements

ObjectivesThe fetal thymus gland has been shown to involute in response to intrauterine infection, and therefore could be used as a non-invasive marker of fetal compartment infection. The objective of this study was to evaluate how accurately 2D ultrasound-derived measurements of the fetal thymus reflect the 3D volume of the gland derived from motion corrected MRI images. Study designA retrospective study was performed using paired ultrasound and MRI datasets from the iFIND project (http://www.ifindproject.com). To obtain 3D volumetry of the thymus gland, T2-weighted single shot turbo spin echo (ssTSE) sequences of the fetal thorax were acquired. Thymus volumes were manually segmented from deformable slice-to-volume reconstructed images. To obtain 2D ultrasound measurements, previously stored fetal cine loops were used and measurements obtained at the 3-vessel-view (3VV) and 3-vessel-trachea view (3VT): anterior-posterior diameter (APD), intrathoracic diameter (ITD), transverse diameter (TD), perimeter and 3-vessel-edge (3VE). Inter-observer and intra-observer reliability (ICC) was calculated for both MRI and ultrasound measurements. Pearson correlation coefficients (PCC) were used to compare 2D-parameters with acceptable ICC to TV. Results38 participants were identified. Adequate visualisation was possible on 37 MRI scans and 31 ultrasound scans. Of the 30 datasets where both MRI and ultrasound data were available, MRI had good interobserver reliability (ICC 0.964) and all ultrasound 3VV 2D-parameters and 3VT 3VE had acceptable ICC (>0.75). Four 2D parameters were reflective of the 3D thymus volume: 3VV TD r = 0.540 (P = 0.002); 3VV perimeter r = 0.446 (P = 0.013); 3VV APD r = 0.435 (P = 0.110) and 3VT TD r = 0.544 (P = 0.002). ConclusionsMRI appeared superior to ultrasound for visualization of the thymus gland and reproducibility of measurements. Three 2D US parameters, 3VV TD, perimeter and 3VT APD, correlated well with TV. Therefore, these represent a more accurate reflection of the true size of the gland than other 2D measurements, where MRI is not available.

Role of Hysteroscopy in Repeated IVF/ICSI – ETs Failure and its Impact Effect on Subsequent Attempts

Background: Recurrent implantation breakdown is characterized as a complication to accomplish a clinical pregnancy subsequent to transfer of 4 or further high-quality embryos in a least amount of 2 in vitro fertilization (IVF) cycles in a woman less than the age of 40. This investigation intended to the responsibility of hysteroscopy in repeated IVF/ICSI failure and its impact on subsequent trials. Methods: Prospective observational study was conducted at assisted reproductive unit in obstetric and gynecology department in Tanta University Hospital from Jan 2018 to Jan 2020. Included 50 infertile patients with more then two IVF/ICSI failure in spite of transmit of as a minimum two better quality embryos. All patients were subjected to 4D US examination with same device (Siemens acuson X300). Results: We found that from the 30 patients who had normal findings in hysteroscopy, 25 had normal findings also in 4DUS, 2 had polyps, 2 had myoma and 1 patient had septum. From the 3 patients who had polyps in hysterectomy, 1 had normal findings in 4DUS, and 2 patients had polyps. Regarding the two patients who had myoma by hysteroscopy, 1 had normal findings in 3DUS and the other had Myoma. While for the 3 considered patients who had septum in hysteroscopy, all had normal findings in 4DUS. Conclusion: According to this study, hysteroscopy is regarded the gold standard not only for envisaging the cervical canal and uterine cavity, however also for healing several types of benign pathologies localized to that area, it is a secure and uncomplicated process.

Minimally invasive management of juvenile cystic adenomyoma: report of three cases

Juvenile cystic adenomyosis (JCA) represents a rare form of focal adenomyosis in young women. To determine safety and effectiveness of minimally invasive surgery (MIS for JCA). Three patients aged 16-30 years old presented with chronic pelvic pain [2016 - 2019]. Hormonal treatment failed in two cases. Cystic lesions in the myometrium (n=2), and the broad ligament (n=1) was detected on transvaginal 2D ultrasound (TV 2D US) and/or magnetic resonance imaging (MRI). The cyst was separate from the endometrium in all the cases, within the myometrium in two patients and in the right broad ligament in one case. The cystic lesions were confirmed on laparoscopy; and laparoscopic excision of the cysts with adequate repair of the myometrial beds were performed in all cases with fertility preservation. Robotic assistance was chosen in one case in an attempt to avoid injury of the fallopian tube based on the cyst location during a previous laparoscopy. The endometrial cavity was entered in one case. Absence of intraoperative complications and relief of presenting symptoms postoperatively. Pathology report confirmed the diagnosis of JCA is all cases. There were no intraoperative complications. All three patients reported relief of their symptoms 6 to 8 months after surgery. No recurrence of the JCA was reported using TV 2D US in all cases. MIS could be the treatment of choice for patients with JCA. The technique described in our study is safe, effective, and easy to master in experienced hands. Value of MIS in treatment of patients with JCA.

112 The association between ovarian maldescent and uterine septum

Maldescended ovaries are rarely encountered and frequently associated with unicornuate uterus and Mϋllerian agenesis. Clinical presentation varies and may mimic other conditions and pathologies; therefore, diagnoses can be challenging. The most common setting of ovarian maldescent is during infertility workup. The aim of this case series study is to increase the awareness of practitioners regarding the association between ovarian maldescent and septate uterus and significant arcuate anomaly in infertile patients or patients with chronic pelvic pain (CPP). Our study is a retrospective case series of four patients; three of whom presented with primary infertility and one with CPP (2014-2018). As part of our basic workup imaging studies were obtained. In all patients it was difficult to visualize the ovaries on trans-vaginal 2D ultrasound (TV 2D US). Hysterosalpingogram in one patient demonstrated fallopian tubes to be elongated and drawn cephalad. Uterine anomalies were suspected on TV 3D US in all four patients. All patients had laparoscopy and hysteroscopy. Hysteroscopy confirmed uterine anomalies, which included one septate uterus, two subseptate uteri, and one significant arcuate anomaly. Septoplasty was performed in all cases. Diagnostic laparoscopy showed patent elongated fallopian tubes and oval-shaped ovaries reaching the pelvic brim in all patients. Three patients had stage 1 endometriosis requiring argon beam coagulation and two required minimal ovariolysis. All three patients who had primary infertility underwent controlled ovarian stimulation (COS) and intrauterine insemination. Monitoring of follicular growth by TV 2D US was sometimes difficult and therefore, trans-abdominal US was also utilized on some occasions. Two patients achieved pregnancy, and both had singleton live birth at term. The remaining patient did not achieve pregnancy after one cycle of COS, and for social reasons discontinued treatment. Our fourth patient who initially presented for CPP and septate uterus was not seeking pregnancy and her symptoms improved after surgery. In conclusion ovarian maldescent is a rare pathology where the ovaries may be found outside the ovarian fossa. Our data suggests that this rare entity can occur in association with septate uterus, subseptate uterus and significant arcuate uterine anomaly. In addition, endometriosis is frequently present in patients with ovarian maldescent. Our experience also suggests a good reproductive outcome after septoplasty and appropriate fertility treatment in such patients. During in-vitro fertilization treatment of patients with ovarian maldescent, trans-abdominal US guided oocyte retrieval may be required if the trans-vaginal US guided retrieval proves to be difficult.

Leveraging voxel-wise segmentation uncertainty to improve reliability in assessment of paediatric dysplasia of the hip.

Estimating uncertainty in predictions made by neural networks is critically important for increasing the trust medical experts have in automatic data analysis results. In segmentation tasks, quantifying levels of confidence can provide meaningful additional information to aid clinical decision making. In recent work, we proposed an interpretable uncertainty measure to aid clinicians in assessing the reliability of developmental dysplasia of the hip metrics measured from 3D ultrasound screening scans, as well as that of the US scan itself. In this work, we propose a technique to quantify confidence in the associated segmentation process that incorporates voxel-wise uncertainty into the binary loss function used in the training regime, which encourages the network to concentrate its training effort on its least certain predictions. We propose using a Bayesian-based technique to quantify 3D segmentation uncertainty by modifying the loss function within an encoder-decoder type voxel labeling deep network. By appending a voxel-wise uncertainty measure, our modified loss helps the network improve prediction uncertainty for voxels that are harder to train. We validate our approach by training a Bayesian 3D U-Net with the proposed modified loss function on a dataset comprising 92 clinical 3D US neonate scans and test on a separate hold-out dataset of 24 patients. Quantitatively, we show that the Dice score of ilium and acetabulum segmentation improves by 5% when trained with our proposed voxel-wise uncertainty loss compared to training with standard cross-entropy loss. Qualitatively, we further demonstrate how our modified loss function results in meaningful reduction of voxel-wise segmentation uncertainty estimates, with the network making more confident accurate predictions. We proposed a Bayesian technique to encode voxel-wise segmentation uncertainty information into deep neural network optimization, and demonstrated how it can be leveraged into meaningful confidence measures to improve the model's predictive performance.

The accuracy of three-dimensional ultrasonography in the diagnosis of Müllerian duct anomalies and its concordance with magnetic resonance imaging

This retrospective study was performed to comparatively evaluate the diagnostic accuracies of three-dimensional ultrasonography (3D-US) and magnetic resonance imaging (MRI) for identification of Müllerian duct anomalies (MDAs). A total of 27 women with suspected MDAs underwent gynaecological examination, 2D-US, 3D-US and MRI, respectively. The MDAs were classified with respect to the European Society of Human Reproduction and Embryology–European Society for Gynaecological Endoscopy (ESHRE/ESGE) and American Society of Reproductive Medicine (ASRM) systems. Based on the ESHRE/ESGE classification, there was a discrepancy for only one patient between US and MRI. Thus, the concordance between US and MRI was 26/27 (96.3%). With respect to ASRM classification, there was a disagreement between MRI and 3D-US in three patients, thus the concordance between MRI and 3D-US was 24/27 (88.9%). To conclude, the 3D-US has a good level of agreement with MRI for recognition of MDAs. Impact Statement What is already known on this subject? Müllerian duct anomalies (MDAs) are relatively common malformations of the female genital tract and they may adversely affect the reproductive potential. The establishment of accurate and timely diagnosis of these malformations is critical to overcome clinical consequences of MDAs. What the results of this study add? The concordance between US and MRI for diagnosis of MDAs based on ESHRE-ESGE classification and ASRM were 96.3% and 88.9%, respectively. These results indicate that 3D US has a satisfactory level of diagnostic accuracy for MDAs and it can be used in conjunction with MRI. Minimisation of diagnostic errors is important to improve reproductive outcome and to avoid unnecessary surgical interventions. What the implications are of these findings for clinical practice and/or further research? Efforts must be spent to eliminate the discrepancies between the clinical and radiological diagnosis of MDAs. Further trials should be implemented for establishment and standardisation of radiological images for identification and classification of MDAs.

Validation of a three dimensional musculoskeletal ultrasound system for use in hands and wrists

Purpose: The effect of chronic synovitis on the progression of osteoarthritis pathology is a relatively new but supported concept in the fields of rheumatology and hand surgery that has been linked to pain and disability in patients. Currently, the gold standard for diagnosing synovitis in the wrists and hands is a clinical examination and the use of 2 dimensional (2D) musculoskeletal ultrasound imaging (US). Major limitations associated with this method are the lack of quantitative measurement tools available and the fact that physicians have to be able to mentally reconstruct the patient’s 3-dimensional anatomy from the 2-dimensional images provided. These limitations can be overcome by using a 3-dimensional ultrasound (3DUS) system to quantitatively measure the size of synovial hypertrophy, effusions, ganglions and osteophytes as well as having a 3D reconstruction of the patient’s hand. The aim of this work is to develop a 3D ultrasound system for imaging the hand and wrist. This paper reports on the development, geometric, and volumetric validation for wrist/hand 3D ultrasound imaging. Methods: A mechatronic wrist/hand imaging system was designed and created to operate with a conventional 2D US imaging system. The scanning system features a motor which contains a single integrated electromagnetic encoder that records the rotational position of the motor. This motor drives a lead screw which in turn moves the transducer, which is submerged in a water bath, along the linear axis to acquire a series of 2D images. The encoder information provides the software with the position of each 2D ultrasound frame. These conventional diagnostic 2D US images are then reconstructed into a 3D US images from a licensed ultrasound machine. The geometric validation of the system was done using a multi-layered monofilament grid phantom with spacings of 10 mm and the Euclidean distances between the wires were measured. The volumetric validation was conducted using a simulated synovial tissue phantom and the images were analyzed using 3D-Slicer to manually segment individual slices in the 3D image. These slices were then interpolated to construct a 3D segmentation of the phantom. This process was repeated over four days to test the intra-rater reliability of the system. The 3D volumes were compared to those found using magnetic resonance imaging (MRI) and the known volume of the agar phantom. Results: The largest error and variance recorded from the geometric validation measurements was 0.0856± 0.0087mm and was acquired from the reconstruction plane. The remaining linear measurement errors were all < 1%. The error between the mean phantom volume measurements and the known synovium phantom volume was 5.99% and 2.43% for the 3D US device and MRI respectively. Linear regression for volumetric measures revealed an R2 value of 0.4736 with a p-value of 0.20, revealing no significant correlation between the measurements made over the course of 4 days for the 3D US device. The percent difference in the volume measurements from the 3D US device and MRI was 0.814%. Conclusions: A mechatronic 3D ultrasound imaging system was developed for the wrist and hand. Successful geometric and volumetric validation demonstrates feasibility for implementation in a clinical setting. Current work evaluates system ability to accurately and quantitatively classify synovitis in first carpometacarpal joint and distal radial injury patients to investigate the relationship between chronic inflammation and the progression of osteoarthritis in patients.

Three-dimensional periodontal investigations using a prototype handheld ultrasound scanner with spatial positioning reading sensor

To demonstrate the feasibility of the 3D ultrasound periodontal tissue reconstruction of the lateral area of a porcine mandible using standard 2D ultrasound equipment and spatial positioning reading sensors. Periodontal 3D reconstructions were performed using a free-hand prototype based on a 2D US scanner and a spatial positioning reading sensor. For automated data processing, deep learning algorithms were implemented and trained using semi-automatically seg-mented images by highly specialized imaging professionals. US probe movement analysis showed that non-parallel 2D frames were acquired during the scanning procedure. Comparing 3 different 3D periodontal reconstructions of the same porcine mandible, the accuracy ranged between 0.179 mm and 0.235 mm. The present study demonstrated the diagnostic potential of 3D reconstruction using a free-hand 2D US scanner with spatial positioning readings. The use of auto-mated data processing with deep learning algorithms makes the process practical in the clinical environment for assessment of periodontal soft tissues.

Quantitative Evaluation of an Automated Cone-Based Breast Ultrasound Scanner for MRI-3D US Image Fusion.

Breast cancer is one of the most diagnosed types of cancer worldwide. Volumetric ultrasound breast imaging, combined with MRI can improve lesion detection rate, reduce examination time, and improve lesion diagnosis. However, to our knowledge, there are no 3D US breast imaging systems available that facilitate 3D US - MRI image fusion. In this paper, a novel Automated Cone-based Breast Ultrasound System (ACBUS) is introduced. The system facilitates volumetric ultrasound acquisition of the breast in a prone position without deforming it by the US transducer. Quality of ACBUS images for reconstructions at different voxel sizes (0.25 and 0.50 mm isotropic) was compared to quality of the Automated Breast Volumetric Scanner (ABVS) (Siemens Ultrasound, Issaquah, WA, USA) in terms of signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and resolution using a custom made phantom. The ACBUS image data were registered to MRI image data utilizing surface matching and the registration accuracy was quantified using an internal marker. The technology was also evaluated in vivo. The phantom-based quantitative analysis demonstrated that ACBUS can deliver volumetric breast images with an image quality similar to the images delivered by a currently commercially available Siemens ABVS. We demonstrate on the phantom and in vivo that ACBUS enables adequate MRI-3D US fusion. To our conclusion, ACBUS might be a suitable candidate for a second-look breast US exam, patient follow-up, and US guided biopsy planning.

Automatic segmentation of ventricular volume by 3D ultrasonography in post haemorrhagic ventricular dilatation among preterm infants

To train, evaluate, and validate the application of a deep learning framework in three-dimensional ultrasound (3D US) for the automatic segmentation of ventricular volume in preterm infants with post haemorrhagic ventricular dilatation (PHVD). We trained a 2D convolutional neural network (CNN) for automatic segmentation ventricular volume from 3D US of preterm infants with PHVD. The method was validated with the Dice similarity coefficient (DSC) and the intra-class coefficient (ICC) compared to manual segmentation. The mean birth weight of the included patients was 1233.1 g (SD 309.4) and mean gestational age was 28.1 weeks (SD 1.6). A total of 152 serial 3D US from 10 preterm infants with PHVD were analysed. 230 ventricles were manually segmented. Of these, 108 were used for training a 2D CNN and 122 for validating the methodology for automatic segmentation. The global agreement for manual versus automated measures in the validation data (n = 122) was excellent with an ICC of 0.944 (0.874–0.971). The Dice similarity coefficient was 0.8 (± 0.01). 3D US based ventricular volume estimation through an automatic segmentation software developed through deep learning improves the accuracy and reduces the processing time needed for manual segmentation using VOCAL. 3D US should be considered a promising tool to help deepen our current understanding of the complex evolution of PHVD.

ACCURACY OF MULTISLICE THREE-DIMENSIONAL ULTRASONOGRAPHY IN PREDICTING SPECIFIC PATHOLOGY OF ADNEXAL MASSES

Background: The MSV mode or Tomographic ultrasound imaging (TUI) allows the simultaneous showing of several longitudinal parallel views for a reference (sagittal, coronal, transverse) plane of the target, its direction and rotation, and magnitude and slice depth and length (0.5 to 5 mm) to be modified by region of interest depending on the size of the object's area of expertise. Objective: To evaluate the diagnostic efficacy of 3D US (multi-slice view) characterizing adnexal masses for prediction of specific pathological nature. Materials and Methods: This prospective study was carried out on 70 patients with age between 15 and 82 years, attending the outpatient clinic of Gynecological Department of Al -Azhar University Hospitals diagnosed as having adnexal mass on conventional 2D sonography. Imaging studies for diagnosis of adnexal masses, 3D US MSV (Multislice view) technique. Diagnostic work-up included an Original 2D ultrasound mass evaluation accompanied by 3D US volume assessment. All masses were evaluated by 3D US MSV. Results: According to the color score in all ovarian masses with solid component, either pure solid or heterogeneous, 11.54% of the masses showed no vascular flow, 46.6 % had mild flow, 7.69 % with moderate flow, and 34.6 % with severe vascular flow. Ovarian masses were classified to benign and malignant by 3D US, and according to their final histopathological analysis. There was a clinical significance relationship for age with the final diagnosis. The test performance of US to detect characterizing adnexal masses as benign or malignant with sensitivity and specificity are 52% and 95.5% respectively. Obese patients with BMI more than 30 with Sensitivity and specificity for US were 52 %, 92.8 % respectively were excluded. Sensitivity of US improved from 52% to 63.6%. Conclusion: MSV 3D US can be useful in the anatomical evaluation of adnexal masses, particularly for the identification of papillary projections in adnexal cysts. This new technique has the potential for expert consultation by sending the image volume to them to do offline analysis, even after finishing the examination with the patient.

ACCURACY OF MULTISLICE THREE-DIMENSIONAL ULTRASONOGRAPHY IN PREDICTING SPECIFIC PATHOLOGY OF ADNEXAL MASSES

Background: The MSV mode or Tomographic ultrasound imaging (TUI) allows the simultaneous showing of several longitudinal parallel views for a reference (sagittal, coronal, transverse) plane of the target, its direction and rotation, and magnitude and slice depth and length (0.5 to 5 mm) to be modified by region of interest depending on the size of the object's area of expertise. Objective: To evaluate the diagnostic efficacy of 3D US (multi-slice view) characterizing adnexal masses for prediction of specific pathological nature. Materials and Methods: This prospective study was carried out on 70 patients with age between 15 and 82 years, attending the outpatient clinic of Gynecological Department of Al -Azhar University Hospitals diagnosed as having adnexal mass on conventional 2D sonography. Imaging studies for diagnosis of adnexal masses, 3D US MSV (Multislice view) technique. Diagnostic work-up included an Original 2D ultrasound mass evaluation accompanied by 3D US volume assessment. All masses were evaluated by 3D US MSV. Results: According to the color score in all ovarian masses with solid component, either pure solid or heterogeneous, 11.54% of the masses showed no vascular flow, 46.6 % had mild flow, 7.69 % with moderate flow, and 34.6 % with severe vascular flow. Ovarian masses were classified to benign and malignant by 3D US, and according to their final histopathological analysis. There was a clinical significance relationship for age with the final diagnosis. The test performance of US to detect characterizing adnexal masses as benign or malignant with sensitivity and specificity are 52% and 95.5% respectively. Obese patients with BMI more than 30 with Sensitivity and specificity for US were 52 %, 92.8 % respectively were excluded. Sensitivity of US improved from 52% to 63.6%. Conclusion: MSV 3D US can be useful in the anatomical evaluation of adnexal masses, particularly for the identification of papillary projections in adnexal cysts. This new technique has the potential for expert consultation by sending the image volume to them to do offline analysis, even after finishing the examination with the patient.