Single Evaluation of Use of a Mixed Reality Headset for Intra-Procedural Image-Guidance during a Mock Laparoscopic Myomectomy on an Ex-Vivo Fibroid Model

Uterine fibroids represent the highest prevalence of benign tumors in women, with reports ranging from 4.5% to 68.6%, with a significant bias towards African American women. For uterine fibroids, a significant decision is determining whether fibroids can be successfully removed using minimally invasive (MI) techniques or their removal requires open surgery. Currently, the standard-of-care for intra-procedural visualization for myomectomies is ultrasound, which has low image quality and requires a specially trained assistant. Currently, the state-of-the-art is to obtain a pre-procedural MRI scan of the patient, which can be used for diagnosis and pre-procedural planning. Although proven incredibly useful pre-procedurally, MRI scans are not often used intra-procedurally due to the inconvenient visualization as 2D slices, which are seen on 2D monitors that do not intuitively convey the depth or orientation of the fibroids, as needed to effectively perform myomectomies. To address this limitation, herein, we present the use of a mixed reality headset (i.e., Microsoft HoloLens 2), as a tool for intra-procedural image-guidance during a mock myomectomy of an ex vivo animal uterus. In this work, we created a patient-specific holographic rendering by performing image segmentation of an MRI scan of a custom-made uterine fibroid animal model. A physician qualitatively assessed the usefulness of the renderings for fibroid localization, as compared to the same visualization on a 2D monitor. In conclusion, the use of mixed reality as an intra-procedural image guidance tool for myomectomies was perceived as a better visualization technique that could lead to improvements in MI approaches and make them accessible to patients from lower socioeconomic populations.

Context-Fused Guidance for Image Captioning Using Sequence-Level Training

Recent image captioning models based on the encoder-decoder framework have achieved remarkable success in humanlike sentence generation. However, an explicit separation between encoder and decoder brings out a disconnection between the image and sentence. It usually leads to a rough image description: the generated caption only contains main instances but neglects additional objects and scenes unexpectedly, which reduces the caption consistency of the image. To address this issue, we proposed an image captioning system within context-fused guidance in this paper. It incorporates regional and global image representation as the compositional visual features to learn the objects and attributes in images. To integrate image-level semantic information, the visual concept is employed. To avoid misleading decoding, a context fusion gate is introduced to calculate the textual context by selectively aggregating the information of visual concept and word embedding. Subsequently, the context-fused image guidance is formulated based on the compositional visual features and textual context. It provides the decoder with informative semantic knowledge. Finally, a captioner with a two-layer LSTM architecture is constructed to generate captions. Moreover, to overcome the exposure bias, we train the proposed model through sequence decision-making. The experiments conducted on the MS COCO dataset show the outstanding performance of our work. The linguistic analysis demonstrates that our model improves the caption consistency of the image.

Insular gliomas and tractographic visualization of the connectome

In this video, the authors present a connectome-guided surgical resection of an insular glioma in a 39-year-old woman. Preoperative study with constrained spherical deconvolution (CSD)–based tractography revealed the surrounding brain connectome architecture around the tumor relevant for safe surgical resection. Connectomic information provided detailed maps of the surrounding language and salience networks, including eloquent white matter fibers and cortical regions, which were visualized intraoperatively with image guidance and artificial intelligence (AI)–based brain mapping software. Microsurgical dissection is presented with detailed discussion of the safe boundaries and angles of resection when entering the insular operculum defined by connectomic information. The video can be found here: https://stream.cadmore.media/r10.3171/2021.10.FOCVID21194

Ultrasound-Guided Treatments for the Painful Shoulder

: Shoulder pain is an extremely common condition. The painful shoulder may be the result of a wide spectrum of underlying pathological conditions, including calcific tendinopathy of the rotator cuff, subacromial-subdeltoid bursitis, acromioclavicular or glenohumeral arthritis, tenosynovitis of the long biceps tendon, rotator cuff lesions, and many other less common conditions. Ultrasound imaging is an effective tool for the diagnosis and also for the image guidance of treatment of the majority of these conditions. Several ultrasound-guided procedures are effective for pain relief, such as percutaneous irrigation, intra-bursal or intra-articular drugs injection, fluid aspiration, neural block. This review article aims to summarize and discuss the most common treatment possibilities with ultrasound guidance for the painful shoulder.