Identification and tracking of vertebrae in ultrasound using deep networks with unsupervised feature learning

Artificial intelligence for ultrasound scanning in regional anaesthesia: a scoping review of the evidence from multiple disciplines

“…In the available sources, a variety of human image interpretation was used as the ground truth. This included a range in the number of humans used in the ground truth from one, 7 , 23 , 24 , 25 , 31 , 32 , 34 , 35 , 36 , 38 , 46 , 49 , 58 , 60 , 67 , 72 , 77 two, 8 , 9 , 19 , 21 , 26 , 30 , 33 , 44 , 61 three, 51 , 52 , 53 , 63 , 65 and up to 15. 48 Some sources cited ‘multiple’ humans, used a plural term (e.g.…”

“…In addition to the varying number of humans contributing to the ground truth in model development, there was variation in the specialty and experience of those contributing. Sources described their ground truth human(s) as a volunteer, 64 doctor/physician, 48 , 52 , 63 , 65 sonographer, 7 , 30 , 31 , 32 , 33 , 34 , 35 , 60 , 75 , 77 clinical professional, 27 , 38 or (regional) anaesthetist. 8 , 17 , 18 , 19 , 20 , 21 , 23 , 25 , 26 , 36 , 44 , 45 , 46 , 49 , 51 , 53 , 54 , 58 , 61 , 62 , 67 , 69 The level of expertise of those providing the ground truth was described as trained, 64 experienced, 21 , 30 , 31 , 32 , 33 , 34 , 46 , 48 , 52 , 65 expert, 7 , 8 , 17 , 18 , 19 , 20 , 25 , ...…”

“…The most common output for models processing CNB ultrasound images was classification/identification of vertebral level, the intervertebral space, or both. 30 , 31 , 32 , 33 , 34 , 75 , 76 , 77 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 93 Distance from the skin to the posterior complex/epidural space was the second most common output of models applied to CNB ultrasound. 7 , 80 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 90 , 91 , 92 , 93 Other functions included identifying the optimal needle insertion point, 29 , 35 , 81 , 88 , 89 , 91 structure identification (via segmentation or bounding boxes), 7 , 35 , 37 or 3D reconstruction of the spine.…”

Follow the Curve: Robotic Ultrasound Navigation With Learning-Based Localization of Spinous Processes for Scoliosis Assessment

“…The dataset was relatively small, consisting of around 1300 images. Methods based on pure deep learning techniques are few and are mainly focused on detecting the spine in sagittal probe orientation; These are typically used for spinal injections where the features appear differently and they are not suitable for scoliosis assessment [9], [10].…”

Follow the Curve: Robotic-Ultrasound Navigation with Learning Based Localization of Spinous Processes for Scoliosis Assessment

“…The input image to this FCN has three channels, where the first channel is an original image, the second is the PS-processed image, and the third is the image resulting from computing the confidence map [6]. Methods based on pure deep learning techniques are few and are mainly focused on detecting the spine in different probe orientations (e.g., sagittal); These are typically used for spinal injections where the features appear different and are not suitable for scoliosis assessment [7], [8].…”