Self-Supervised 3-D Semantic Representation Learning for Vision-and-Language Navigation.

In vision-and-language navigation (VLN) tasks, most current methods primarily utilize RGB images, overlooking the rich 3-D semantic data inherent to environments. To rectify this, we introduce a novel VLN framework that integrates 3-D semantic information into the navigation process. Our approach features a self-supervised training scheme that incorporates voxel-level 3-D semantic reconstruction to create a detailed 3-D semantic representation. A key component of this framework is a pretext task focused on region queries, which determines the presence of objects in specific 3-D areas. Following this, we devise an long short-term memory (LSTM)-based navigation model that is trained using our 3-D semantic representations. To maximize the utility of these 3-D semantic representations, we implement a cross-modal distillation strategy. This strategy encourages the RGB model's outputs to emulate those from the 3-D semantic feature network, enabling the concurrent training of both branches to merge RGB and 3-D semantic data effectively. Comprehensive evaluations on both the R2R and R4R datasets reveal that our method significantly enhances performance in VLN tasks.

Radiolocalized sentinel lymph node biopsy in squamous cell carcinoma of the oral cavity and analysis of various parameters.

BACKGROUND: Sentinel lymph node (SLN) biopsy seems to be a method that solves the problem of neck management with oral squamous cell carcinoma. Using blue dye methods for detection of neck SLNs from the surface of the operative field seems difficult; therefore, we used radiolocalization alone to detect and extract sentinel nodes. Aside from the various histological and clinical parameters examined in this procedure, we also determined whether they had any clinical significance in relation to the detection of SLNs during the operation. METHODS: Enrolled subjects had preoperative clinical N0 stage squamous cell carcinoma of the oral cavity and had received an unfiltered (99m)Tc sulfur colloid peritumoral injection. Localization of the SLNs was performed by using lymphoscintigraphy and a handheld gamma probe. RESULTS: In total, 28 oral squamous cell carcinoma patients were included in this prospective study. Sixty-four SLNs in 27 patients were identified by this method. The identification rate was 96.4%. No false-negative predictions of SLN were noted among any of the patients studied. The numbers of the SLNs found during the operation were larger in patients with positive findings than those with negative findings (P < .05 by the Mann-Whitney U-test). CONCLUSIONS: SLN radiolocalization provided an acceptable identification rate. The cases of positive findings for metastasis seemed to statistically have more SLNs than did those with negative findings, but more evidence is needed to prove this point. Therefore, SLN biopsies for extracting all possible high-risk nodes may be conducive for oral squamous cell carcinoma surgery.