OMERACT validation of a deep learning algorithm for automated absolute quantification of knee joint effusion versus manual semi-quantitative assessment.

OBJECTIVE: To begin evaluating deep learning (DL)-automated quantification of knee joint effusion-synovitis via the OMERACT filter. METHODS: A DL algorithm previously trained on Osteoarthritis Initiative (OAI) knee MRI automatically quantified effusion volume in MRI of 53 OAI subjects, which were also scored semi-quantitatively via KIMRISS and MOAKS by 2-6 readers. RESULTS: DL-measured knee effusion correlated significantly with experts' assessments (Kendall's tau 0.34-0.43) CONCLUSION: The close correlation of automated DL knee joint effusion quantification to KIMRISS manual semi-quantitative scoring demonstrated its criterion validity. Further assessments of discrimination and truth vs. clinical outcomes are still needed to fully satisfy OMERACT filter requirements.

Single Image based Super Resolution Ultrasound Imaging Using Residual Learning of Wavelet Features.

The generation of super resolution ultrasound images from the low-resolution (LR) brightness mode (B-mode) images acquired by the portable point of care ultrasound systems has been of sufficient interest in the recent past. With the advancements in deep learning, there have been numerous attempts in this direction. However, all the approaches have been concentrated on employing the direct image as the input to the neural network. In this work, a stationary wavelet (SWT) decomposition is employed to extract the features from the input LR image which is passed through a modified residual network and the learned features are combined using the inverse SWT to reconstruct the high resolution (HR) image at a 4× scale factor. The proposed approach when compared to the state-of-the art approaches, results in an improved high resolution reconstruction.Clinical relevance- The proposed approach will enable the generation of high-resolution images from portable ultrasound systems, allowing for easier interpretation and faster diagnostics in primary care settings.

2D/3D ultrasound diagnosis of pediatric distal radius fractures by human readers vs artificial intelligence.

Wrist trauma is common in children and generally requires radiography for exclusion of fractures, subjecting children to radiation and long wait times in the emergency department. Ultrasound (US) has potential to be a safer, faster diagnostic tool. This study aimed to determine how reliably US could detect distal radius fractures in children, to contrast the accuracy of 2DUS to 3DUS, and to assess the utility of artificial intelligence for image interpretation. 127 children were scanned with 2DUS and 3DUS on the affected wrist. US scans were then read by 7 blinded human readers and an AI model. With radiographs used as the gold standard, expert human readers obtained a mean sensitivity of 0.97 and 0.98 for 2DUS and 3DUS respectively. The AI model sensitivity was 0.91 and 1.00 for 2DUS and 3DUS respectively. Study data suggests that 2DUS is comparable to 3DUS and AI diagnosis is comparable to human experts.

Self-supervised-RCNN for medical image segmentation with limited data annotation.

Many successful methods developed for medical image analysis based on machine learning use supervised learning approaches, which often require large datasets annotated by experts to achieve high accuracy. However, medical data annotation is time-consuming and expensive, especially for segmentation tasks. To overcome the problem of learning with limited labeled medical image data, an alternative deep learning training strategy based on self-supervised pretraining on unlabeled imaging data is proposed in this work. For the pretraining, different distortions are arbitrarily applied to random areas of unlabeled images. Next, a Mask-RCNN architecture is trained to localize the distortion location and recover the original image pixels. This pretrained model is assumed to gain knowledge of the relevant texture in the images from the self-supervised pretraining on unlabeled imaging data. This provides a good basis for fine-tuning the model to segment the structure of interest using a limited amount of labeled training data. The effectiveness of the proposed method in different pretraining and fine-tuning scenarios was evaluated based on the Osteoarthritis Initiative dataset with the aim of segmenting effusions in MRI datasets of the knee. Applying the proposed self-supervised pretraining method improved the Dice score by up to 18% compared to training the models using only the limited annotated data. The proposed self-supervised learning approach can be applied to many other medical image analysis tasks including anomaly detection, segmentation, and classification.

Remote diagnostic imaging using artificial intelligence for diagnosing hip dysplasia in infants: Results from a mixed-methods feasibility pilot study.

Objectives: Infant hip dysplasia or Developmental Dysplasia of the Hip (DDH) occurs in 1-2% of births worldwide and leads to hip arthritis if untreated. We sought to evaluate the feasibility of implementing an artificial intelligence-enhanced portable ultrasound tool for infant hip dysplasia (DDH) screening in primary care, through determining its effectiveness in practice and evaluating patient and provider feedback. Methods: A US-FDA-cleared artificial intelligence (AI) screening device for DDH (MEDO-Hip) was added to routine well-child visits from age 6 to 10 weeks. A total of 306 infants were screened during a 1-year pilot study within three family medicine clinics in Alberta, Canada. Patient and provider satisfaction were quantified using the System Usability Survey (SUS), while provider perceptions were further investigated through semi-structured interviews. Results: Provider and user surveys commonly identified best features of the tool as immediate diagnosis, offering reassurance/knowledge and avoiding travel, and noted technical glitches most frequently as a barrier. A total of 369 scans of 306 infants were performed from Feb 1, 2021 until Mar 31, 2022. Eighty percent of hips scanned were normal on initial scans, 14% of scans required a follow-up study in the primary care clinic, and DDH cases were identified and treated at the expected 2% rate (6 infants). Conclusions: It is feasible to implement a point-of-care ultrasound AI screening tool in primary care to screen for infants with DDH. Beyond improved screening and detection, this innovation was well accepted by patients and fee-for-service providers with a culture and history of innovation.

AI aided workflow for hip dysplasia screening using ultrasound in primary care clinics.

Developmental dysplasia of the hip (DDH) is a common cause of premature osteoarthritis. This osteoarthritis can be prevented if DDH is detected by ultrasound and treated in infancy, but universal DDH screening is generally not cost-effective due to the need for experts to perform the scans. The purpose of our study was to evaluate the feasibility of having non-expert primary care clinic staff perform DDH ultrasound using handheld ultrasound with artificial intelligence (AI) decision support. We performed an implementation study evaluating the FDA-cleared MEDO-Hip AI app interpreting cine-sweep images obtained from handheld Philips Lumify probe to detect DDH. Initial scans were done by nurses or family physicians in 3 primary care clinics, trained by video, powerpoint slides and brief in-person. When the AI app recommended follow-up (FU), we first performed internal FU by a sonographer using the AI app; cases still considered abnormal by AI were referred to pediatric orthopedic clinic for assessment. We performed 369 scans in 306 infants. Internal FU rates were initially 40% for nurses and 20% for physicians, declining steeply to 14% after ~ 60 cases/site: 4% technical failure, 8% normal at sonographer FU using AI, and 2% confirmed DDH. Of 6 infants referred to pediatric orthopedic clinic, all were treated for DDH (100% specificity); 4 had no risk factors and may not have otherwise been identified. Real-time AI decision support and a simplified portable ultrasound protocol enabled lightly trained primary care clinic staff to perform hip dysplasia screening with FU and case detection rates similar to costly formal ultrasound screening, where the US scan is performed by a sonographer and interpreted by a radiologist/orthopedic surgeon. This highlights the potential utility of AI-supported portable ultrasound in primary care.

Unsupervised landmark detection and classification of lung infection using transporter neural networks.

Supervised deep learning techniques have been very popular in medical imaging for various tasks of classification, segmentation, and object detection. However, they require a large number of labelled data which is expensive and requires many hours of careful annotation by experts. In this paper, an unsupervised transporter neural network framework with an attention mechanism is proposed to automatically identify relevant landmarks with applications in lung ultrasound (LUS) imaging. The proposed framework identifies key points that provide a concise geometric representation highlighting regions with high structural variation in the LUS videos. In order for the landmarks to be clinically relevant, we have employed acoustic propagation physics driven feature maps and angle-controlled Radon Transformed frames at the input instead of directly employing the gray scale LUS frames. Once the landmarks are identified, the presence of these landmarks can be employed for classification of the given frame into various classes of severity of infection in lung. The proposed framework has been trained on 130 LUS videos and validated on 100 LUS videos acquired from multiple centres at Spain and India. Frames were independently assessed by experts to identify clinically relevant features such as A-lines, B-lines, and pleura in LUS videos. The key points detected showed high sensitivity of 99% in detecting the image landmarks identified by experts. Also, on employing for classification of the given lung image into normal and abnormal classes, the proposed approach, even with no prior training, achieved an average accuracy of 97% and an average F1-score of 95% respectively on the task of co-classification with 3-fold cross-validation.

Unsupervised Multi-Latent Space RL Framework for Video Summarization in Ultrasound Imaging.

The COVID-19 pandemic has highlighted the need for a tool to speed up triage in ultrasound scans and provide clinicians with fast access to relevant information. To this end, we propose a new unsupervised reinforcement learning (RL) framework with novel rewards to facilitate unsupervised learning by avoiding tedious and impractical manual labelling for summarizing ultrasound videos. The proposed framework is capable of delivering video summaries with classification labels and segmentations of key landmarks which enhances its utility as a triage tool in the emergency department (ED) and for use in telemedicine. Using an attention ensemble of encoders, the high dimensional image is projected into a low dimensional latent space in terms of: a) reduced distance with a normal or abnormal class (classifier encoder), b) following a topology of landmarks (segmentation encoder), and c) the distance or topology agnostic latent representation (autoencoders). The summarization network is implemented using a bi-directional long short term memory (Bi-LSTM) which utilizes the latent space representation from the encoder. Validation is performed on lung ultrasound (LUS), that typically represent potential use cases in telemedicine and ED triage acquired from different medical centers across geographies (India and Spain). The proposed approach trained and tested on 126 LUS videos showed high agreement with the ground truth with an average precision of over 80% and average F1 score of well over 44 ±1.7 %. The approach resulted in an average reduction in storage space of 77% which can ease bandwidth and storage requirements in telemedicine.

Normal variation of infant hip development : patterns revealed by 3D ultrasound.

AIMS: Studies of infant hip development to date have been limited by considering only the changes in appearance of a single ultrasound slice (Graf's standard plane). We used 3D ultrasound (3DUS) to establish maturation curves of normal infant hip development, quantifying variation by age, sex, side, and anteroposterior location in the hip. METHODS: We analyzed 3DUS scans of 519 infants (mean age 64 days (6 to 111 days)) presenting at a tertiary children's hospital for suspicion of developmental dysplasia of the hip (DDH). Hips that did not require ultrasound follow-up or treatment were classified as 'typically developing'. We calculated traditional DDH indices like α angle (αSP), femoral head coverage (FHCSP), and several novel indices from 3DUS like the acetabular contact angle (ACA) and osculating circle radius (OCR) using custom software. RESULTS: α angle, FHC, and ACA indices increased and OCR decreased significantly by age in the first four months, mean αSP rose from 62.2° (SD 5.7°) to 67.3° (SD 5.2°) (p < 0.001) in one- to eight- and nine- to 16-week-old infants, respectively. Mean αSP and mean FHCSP were significantly, but only slightly, lower in females than in males. There was no statistically significant difference in DDH indices observed between left and right hip. All 3DUS indices varied significantly between anterior and posterior section of the hip. Mean 3D indices of α angle and FHC were significantly lower anteriorly than posteriorly: αAnt = 58.2° (SD 6.1°), αPost = 63.8° (SD 6.3°) (p < 0.001), FHCAnt = 43.0 (SD 7.4), and FHCPost = 55.4° (SD 11.2°) (p < 0.001). Acetabular rounding measured byOCR indices was significantly greater in the anterior section of the hip (p < 0.001). CONCLUSION: We used 3DUS to show that hip shape and normal growth pattern vary significantly between anterior and posterior regions, by magnitudes similar to age-related changes. This highlights the need for careful selection of the Graf plane during 2D ultrasound examination. Whole-joint evaluation by obtaining either 3DUS or manual 'sweep' video images provides more comprehensive DDH assessment.Cite this article: Bone Jt Open 2022;3(11):913-923.

Domain-aware contrastive learning for ultrasound hip image analysis.

Early diagnosis of Developmental Dysplasia of Hip (DDH) using ultrasound can result in simpler and more effective treatment options. Handheld ultrasound probes are ideally suited for such screening due to their low cost and portability. However, images from the pocket-sized probes are of lower quality than conventional probes. Image quality can be enhanced by image translation techniques that generate a pseudo-image mimicking the image quality of conventional probes. This can also help in generalizing the performance of AI-based automatic interpretation techniques to multiple probes. We develop a new domain-aware contrastive unpaired translation (D-CUT) technique for translating between images acquired from different ultrasound probes. Our approach embeds a Bone Probability Map (BPM) as part of the loss function which enforces higher structural similarity around bony regions in the image. Using the D-CUT model we translated 575 images acquired from a Philips Lumify handheld probe to generate pseudo-3D ultrasound (3DUS) images similar (Fréchet Inception Distance = 92) to those acquired from a conventional ultrasound probe (Philips iU22). The pseudo-3DUS images showed high structural similarity (SSIM = 0.68, Cosine Similarity = 0.65) with the original images and improved the contrast around the bony regions. This study establishes the feasibility of using D-CUT to improve the quality of data acquired from handheld ultrasound probes. Among other potential applications, clinical use of this tool could result in wider use of ultrasound for DDH screening programs.

Improved-Mask R-CNN: Towards an accurate generic MSK MRI instance segmentation platform (data from the Osteoarthritis Initiative).

INTRODUCTION: Objective assessment of osteoarthritis (OA) Magnetic Resonance Imaging (MRI) scans can address the limitations of the current OA assessment approaches. Detecting and extracting bone, cartilage, and joint fluid is a necessary component for the objective assessment of OA, which helps to quantify tissue characteristics such as volume and thickness. Many algorithms, based on Artificial Intelligence (AI), have been proposed over recent years for segmenting bone and soft tissues. Most of these segmentation methods suffer from the class imbalance problem, can't differentiate between the same anatomic structure, or do not support segmenting different rang of tissue sizes. Mask R-CNN is an instance segmentation framework, meaning it segments and distinct each object of interest like different anatomical structures (e.g. bone and cartilage) using a single model. In this study, the Mask R-CNN architecture was deployed to address the need for a segmentation method that is applicable to use for different tissue scales, pathologies, and MRI sequences associated with OA, without having a problem with imbalanced classes. In addition, we modified the Mask R-CNN to improve segmentation accuracy around instance edges. METHODS: A total of 500 adult knee MRI scans from the publicly available Osteoarthritis Initiative (OAI), and 97 hip MRI scans from adults with symptomatic hip OA, evaluated by two readers, were used for training and validating the network. Three specific modifications to Mask R-CNN yielded the improved-Mask R-CNN (iMaskRCNN): an additional ROIAligned block, an extra decoder block in the segmentation header, and connecting them using a skip connection. The results were evaluated using Hausdorff distance, dice score for bone and cartilage segmentation, and differences in detected volume, dice score, and coefficients of variation (CoV) for effusion segmentation. RESULTS: The iMaskRCNN led to improved bone and cartilage segmentation compared to Mask RCNN as indicated with the increase in dice score from 95% to 98% for the femur, 95-97% for the tibia, 71-80% for the femoral cartilage, and 81-82% for the tibial cartilage. For the effusion detection, the dice score improved with iMaskRCNN 72% versus Mask R-CNN 71%. The CoV values for effusion detection between Reader1 and Mask R-CNN (0.33), Reader1 and iMaskRCNN (0.34), Reader2 and Mask R-CNN (0.22), Reader2 and iMaskRCNN (0.29) are close to CoV between two readers (0.21), indicating a high agreement between the human readers and both Mask R-CNN and iMaskRCNN. CONCLUSION: Mask R-CNN and iMaskRCNN can reliably and simultaneously extract different scale articular tissues involved in OA, forming the foundation for automated assessment of OA. The iMaskRCNN results show that the modification improved the network performance around the edges.

Interobserver Variability of Hip Dysplasia Indices on Sweep Ultrasound for Novices, Experts, and Artificial Intelligence.

BACKGROUND: Ultrasound for developmental dysplasia of the hip (DDH) is challenging for nonexperts to perform and interpret. Recording "sweep" images allows more complete hip assessment, suitable for automation by artificial intelligence (AI), but reliability has not been established. We assessed agreement between readers of varying experience and a commercial AI algorithm, in DDH detection from infant hip ultrasound sweeps. METHODS: We selected a full spectrum of poor-to-excellent quality images and normal to severe dysplasia, in 240 hips (120 single 2-dimensional images, 120 sweeps). For 12 readers (radiologists, sonographers, clinicians and researchers; 3 were DDH subspecialists), and a ultrasound-FDA-cleared AI software package (Medo Hip), we calculated interobserver reliability for alpha angle measurements by intraclass correlation coefficient (ICC2,1) and for DDH classification by Randolph Kappa. RESULTS: Alpha angle reliability was high for AI versus subspecialists (ICC=0.87 for sweeps, 0.90 for single images). For DDH diagnosis from sweeps, agreement was high between subspecialists (kappa=0.72), and moderate for nonsubspecialists (0.54) and AI (0.47). Agreement was higher for single images (kappa=0.80, 0.66, 0.49). AI reliability deteriorated more than human readers for the poorest-quality images. The agreement of radiologists and clinicians with the accepted standard, while still high, was significantly poorer for sweeps than 2D images (P<0.05). CONCLUSIONS: In a challenging exercise representing the wide spectrum of image quality and reader experience seen in real-world hip ultrasound, agreement on DDH diagnosis from easily obtained sweeps was only slightly lower than from single images, likely because of the additional step of selecting the best image. AI performed similarly to a nonsubspecialist human reader but was more affected by low-quality images.

A New Semi-automated Algorithm for Volumetric Segmentation of the Left Ventricle in Temporal 3D Echocardiography Sequences.

PURPOSE: Echocardiography is commonly used as a non-invasive imaging tool in clinical practice for the assessment of cardiac function. However, delineation of the left ventricle is challenging due to the inherent properties of ultrasound imaging, such as the presence of speckle noise and the low signal-to-noise ratio. METHODS: We propose a semi-automated segmentation algorithm for the delineation of the left ventricle in temporal 3D echocardiography sequences. The method requires minimal user interaction and relies on a diffeomorphic registration approach. Advantages of the method include no dependence on prior geometrical information, training data, or registration from an atlas. RESULTS: The method was evaluated using three-dimensional ultrasound scan sequences from 18 patients from the Mazankowski Alberta Heart Institute, Edmonton, Canada, and compared to manual delineations provided by an expert cardiologist and four other registration algorithms. The segmentation approach yielded the following results over the cardiac cycle: a mean absolute difference of 1.01 (0.21) mm, a Hausdorff distance of 4.41 (1.43) mm, and a Dice overlap score of 0.93 (0.02). CONCLUSION: The method performed well compared to the four other registration algorithms.

Impact of scan quality on AI assessment of hip dysplasia ultrasound.

AIMS: Early diagnosis of developmental dysplasia of the hip (DDH) using ultrasound (US) is safe, effective and inexpensive, but requires high-quality scans. The effect of scan quality on diagnostic accuracy is not well understood, especially as artificial intelligence (AI) begins to automate such diagnosis. In this paper, we developed a 10-point scoring system for reporting DDH US scan quality, evaluated its inter-rater agreement and examined its effect on automated assessment by an AI system-MEDO-Hip. METHODS: Scoring was based on iliac wing straightness and angulation; visibility of labrum, os ischium and femoral head; motion; and other artifacts. Four readers from novice to expert separately scored the quality of 107 scans with this 10-point scale and with holistic grading on a scale of 1-5. MEDO-Hip interpreted the same scans, providing a diagnostic category or identifying the scan as uninterpretable. RESULTS: Inter-rater agreement for the 10-point scale was significantly higher than holistic scoring ICC 0.68 vs 0.93, p < 0.05. Inter-rater agreement on the categorisation of individual features, by Cohen's kappa, was highest for os ischium (0.67 ± 0.06), femoral head (0.65 ± 0.07) and iliac wing (0.49 ± 0.12) indices, and lower for the presence of labrum (0.21 ± 0.19). MEDO-Hip interpreted all images of a quality > 7 and flagged 13/107 as uninterpretable. These were low-quality images (3 ± 1.2 vs. 7 ± 1.8 in others, p < 0.05), with poor visualization of the os ischium and noticeable motion. AI accuracy in cases with quality scores < = 7 was 57% vs. 89% on other cases, p < 0.01. CONCLUSION: This study validates that our scoring system reliably characterises scan quality, and identifies cases likely to be misinterpreted by AI. This could lead to more accurate use of AI in DDH diagnosis by flagging low-quality scans likely to provide poor diagnosis up front.

Integrating User-Input into Deep Convolutional Neural Networks for Thyroid Nodule Segmentation.

Delineation of thyroid nodule boundaries is necessary for cancer risk assessment and accurate categorization of nodules. Clinicians often use manual or bounding-box approach for nodule assessment which leads to subjective results. Consequently, agreement in thyroid nodule categorization is poor even among experts. Computer-aided diagnosis systems could reduce this variability by minimizing the extent of user interaction and by providing precise nodule segmentations. In this study, we present a novel approach for effective thyroid nodule segmentation and tracking using a single user click on the region of interest. When a user clicks on an ultrasound sweep, our proposed model can predict nodule segmentation over the entire sequence of frames. Quantitative evaluations show that the proposed method out-performs the bounding box approach in terms of the dice score on a large dataset of 372 ultrasound images. The proposed approach saves expert time and reduces the potential variability in thyroid nodule assessment. The proposed one-click approach can save clinicians time required for annotating thyroid nodules within ultrasound images/sweeps. With minimal user interaction we would be able to identify the nodule boundary which can further be used for volumetric measurement and characterization of the nodule. This approach can also be extended for fast labeling of large thyroid imaging datasets suitable for training machine-learning based algorithms.

Automatic Assessment Of Hip Effusion From MRI.

Joint effusion is a hallmark of osteoarthritis (OA) associated with stiffness, and may relate to pain, disability, and long-term outcomes. However, it is difficult to quantify accurately. We propose a new Deep Learning (DL) approach for automatic effusion assessment from Magnetic Resonance Imaging (MRI) using volumetric quantification measures (VQM). We developed a new multiplane ensemble convolutional neural network (CNN) approach for 1) localizing bony anatomy and 2) detecting effusion regions. CNNs were trained on femoral head and effusion regions manually segmented from 3856 images (63 patients). Upon validation on a non-overlapping set of 2040 images (34 patients) DL showed high agreement with ground-truth in terms of Dice score (0.85), sensitivity (0.86) and precision (0.83). Agreement of VQM per-patient was high for DL vs experts in term of Intraclass correlation coefficient (ICC)= 0.88[0.80,0.93]. We expect this technique to reduce inter-observer variability in effusion assessment, reducing expert time and potentially improving the quality of OA care.Clinical Relevance- Our technique for automatic assessment of hip MRI can be used for volumetric measurement of effusion. We expect this to reduce variability in OA biomarker assessment and provide more reliable indicators for disease progression.

Thyroid Nodule Segmentation and Classification Using Deep Convolutional Neural Network and Rule-based Classifiers.

Thyroid cancer has a high prevalence all over the world. Accurate thyroid nodule diagnosis can lead to effective treatment and decrease the mortality rate. Ultrasound imaging is a safe, portable, and inexpensive tool for thyroid nodule monitoring. However, the widespread use of ultrasound has also resulted in over-diagnosis and over-treatment of nodules. There is also large variability in the assessment and characterization of nodules. Thyroid nodule classification requires precise delineation of the nodule boundary which is tedious and time- consuming. Automatic segmentation of nodule boundaries is highly desirable, however, it is challenging due to the wide range of nodule appearances, shapes, and sizes. In this study, we propose an end-to-end pipeline for nodule segmentation and classification. A residual dilated UNet (resDUnet) model is proposed for nodule segmentation. The output of resDUnet is fed to two rule-based classifiers to categorize the composition and echogenicity of the segmented nodule. We evaluate our segmentation method on a large dataset of 352 ultrasound images reviewed by a certified radiologist. When compared with ground-truth, resDUnet gives a higher Dice score than the standard UNet (82% vs. 81%). Our method requires minimal user interaction and it is robust to reasonable variations in the user-specified region-of-interest. We expect the proposed method to reduce variability in thyroid nodule assessment which results in more efficient and cost-effective monitoring of thyroid cancer.

MRI Knee Domain Translation for Unsupervised Segmentation By CycleGAN (data from Osteoarthritis initiative (OAI)).

Accurate quantification of bone and cartilage features is the key to efficient management of knee osteoarthritis (OA). Bone and cartilage tissues can be accurately segmented from magnetic resonance imaging (MRI) data using supervised Deep Learning (DL) methods. DL training is commonly conducted using large datasets with expert-labeled annotations. DL models perform better if distributions of testing data (target domains) are close to those of training data (source domains). However, in practice, data distributions of images from different MRI scanners and sequences are different and DL models need to re-trained on each dataset separately. We propose a domain adaptation (DA) framework using the CycleGAN model for MRI translation that would aid in unsupervised MRI data segmentation. We have validated our pipeline on five scans from the Osteoarthritis Initiative (OAI) dataset. Using this pipeline, we translated TSE Fat Suppressed MRI sequences to pseudo-DESS images. An improved MaskRCNN (IMaskRCNN) instance segmentation network trained on DESS was used to segment cartilage and femoral head regions in TSE Fat Suppressed sequences. Segmentations of the I-MaskRCNN correlated well with approximated manual segmentation obtained from nearest DESS slices (DICE = 0.76) without the need for retraining. We anticipate this technique will aid in automatic unsupervised assessment of knee MRI using commonly acquired MRI sequences and save experts' time that would otherwise be required for manual segmentation.Clinical relevance- This technique paves the way to automatically convert one MRI sequence to its equivalent as if acquired by a different protocol or different magnet, facilitating robust, hardware-independent automated analysis. For example, routine clinically acquired knee MRI could be converted to high-resolution high-contrast images suitable for automated detection of cartilage defects.

vid-SAMGRAH: A PyTorch framework for multi-latent space reinforcement learning driven video summarization in ultrasound imaging.

The COVID-19 pandemic has accelerated the need for automatic triaging and summarization of ultrasound videos for fast access to pathologically relevant information in the Emergency Department and lowering resource requirements for telemedicine. In this work, a PyTorch based unsupervised reinforcement learning methodology which incorporates multi feature fusion to output classification labels, segmentation maps and summary videos for lung ultrasound is presented. The use of unsupervised training eliminates tedious manual labeling of key-frames by clinicians opening new frontiers in scalability in training using unlabeled or weakly labeled data. Our approach was benchmarked against expert clinicians from different geographies displaying superior Precision and F1 scores (over 80% and 44%).

Multi-View 3-D Fusion Echocardiography: Enhancing Clinical Feasibility with a Novel Processing Technique.

A novel system for fusing 3-D echocardiography data sets from complementary acoustic windows was evaluated in 12 healthy volunteers and 12 patients with heart failure. We hypothesized that 3-D fusion would enable 3-D echocardiography in patients with limited acoustic windows. At least nine 3-D data sets were recorded, while three infrared cameras tracked the position and orientation of the transducer and chest respiratory movements. Corresponding 2-D planes of the fused 3-D data sets and of single-view 3-D data sets were assessed for image quality and compared with measurements of left ventricular function obtained with contrast 2-D echocardiography. The signal-to-noise ratio in accurately fused 3-D echocardiography recordings improved by 55% in systole (p < 0.001) and 47% in diastole (p < 0.00001) compared with the apical single-view recordings. The 3-D data sets acquired during short breath holds were successfully fused in 11 of 12 patients. The improvement in endocardial border definition (from 11.7 ± 6.0 to 24.0 ± 3.3, p < 0.01) enabled quantitative assessment of left ventricular function in 10 patients, with no significant difference in ejection fraction compared with contrast 2-D echocardiography. In patients with heart failure and limited acoustic windows, the novel fusion protocol provides 3-D data sets suitable for quantitative analysis of left ventricular function.