An ingenious deep learning approach for pressure injury depth evaluation with limited data.

BACKGROUND: The development of models using deep learning (DL) to assess pressure injuries from wound images has recently gained attention. Creating enough supervised data is important for improving performance but is time-consuming. Therefore, the development of models that can achieve high performance with limited supervised data is desirable. MATERIALS AND METHODS: This retrospective observational study utilized DL and included patients who received medical examinations for sacral pressure injuries between February 2017 and December 2021. Images were labeled according to the DESIGN-R® classification. Three artificial intelligence (AI) models for assessing pressure injury depth were created with a convolutional neural network (Categorical, Binary, and Combined classification models) and performance was compared among the models. RESULTS: A set of 414 pressure injury images in five depth stages (d0 to D4) were analyzed. The Combined classification model showed superior performance (F1-score, 0.868). The Categorical classification model frequently misclassified d1 and d2 as d0 (d0 Precision, 0.503), but showed high performance for D3 and D4 (F1-score, 0.986 and 0.966, respectively). The Binary classification model showed high performance in differentiating between d0 and d1-D4 (F1-score, 0.895); however, performance decreased with increasing number of evaluation steps. CONCLUSION: The Combined classification model displayed superior performance without increasing the supervised data, which can be attributed to use of the high-performance Binary classification model for initial d0 evaluation and subsequent use of the Categorical classification model with fewer evaluation steps. Understanding the unique characteristics of classification methods and deploying them appropriately can enhance AI model performance.

Factors Associated with Complications after Total Pharyngo-Laryngo-Esophagectomy and Free Jejunal Flap Reconstruction.

INTRODUCTION: Free jejunal flap (FJF) reconstruction is a standard procedure for pharyngeal and cervical esophageal defects resulting from head and neck cancer resection. However, improvements in patients' quality of life after surgery require a further statistical approach. METHODS: An observational, retrospective, multivariate analysis was designed to report the incidence of postoperative complications and their association with clinical factors in 101 patients who underwent total pharyngo-laryngo-esophagectomy and FJF reconstruction for head and neck cancer at a university hospital between January 2007 and December 2020. RESULTS: Postoperative complications were observed in 69% of patients. In the reconstructive site, anastomotic leak, observed in 8% of patients was associated with vascular anastomosis in the external jugular vein system (age-adjusted odds ratio [OR]: 9.05, p = 0.044) and anastomotic stricture, observed in 11% of patients was associated with postoperative radiotherapy (age-adjusted OR: 12.60, p = 0.02). Cervical skin flap necrosis was the most common complication (34%) and was associated with vascular anastomosis on the right cervical side (age- and sex-adjusted OR: 4.00, p = 0.005). CONCLUSION: Although FJF reconstruction is a useful procedure, 69% of patients suffer a postoperative complication. We suppose that anastomotic leak is related to the low blood flow resistance of the FJF and inadequate drainage of the external jugular venous system, and anastomotic stricture is related to the vulnerability of the intestinal tissue to radiation. Furthermore, we hypothesized that the location of the vascular anastomosis may affect the mesenteric location of the FJF and the dead space in the neck, leading to the development of cervical skin flap necrosis. These data contribute to increasing our knowledge about postoperative complications related to FJF reconstruction.

Reconstruction for Bilateral Internal Jugular Vein Perfusion Disruption.

When resecting the internal jugular veins bilaterally in surgery for head and neck cancer, it is necessary to perform neck dissection in two stages or to reconstruct the internal jugular veins in one stage. Reconstruction of the internal jugular vein using grafting or direct anastomosis to the external jugular vein have both been reported. We report the case of a 53-year-old man with accidental injury to the left internal jugular vein after resection of the right internal jugular vein for supraglottic cancer. The left internal jugular vein was damaged near the inflow of the subclavian vein, making vein grafting difficult. Therefore, internal jugular venous return was reestablished by end-to-side anastomosis of the left internal jugular vein to the left external jugular vein system. In this surgical procedure, by incising the internal jugular vein obliquely, it was not necessary to match the calibers of the internal jugular vein and the external jugular vein system, and a smooth hemodynamic body was reconstructed. In addition, we were able to reconstruct the internal jugular vein while preserving blood flow in the external jugular vein system. End-to-side anastomosis of the internal jugular vein to the external jugular system is an option for internal jugular vein reconstruction.

Comparison of Antera 3D® and TcPO2 for Evaluation of Blood Flow in Skin.

Background: There is a need for quick skin blood flow tests that can be performed in the wound healing field. Antera 3D® is a compact scanner using multispectral imaging. It can perform quick assessment of skin conditions. The purpose of the present study was to investigate the ability of the Antera 3D® to assess skin blood flow in comparison with transcutaneous partial pressure of oxygen (TcPO2) measurements. Methods: This study was conducted on 13 patients with a history of lower extremity ulcers. Measurements of hemoglobin average level (hereafter, Hb score) measured by Antera 3D® and TcPO2 measured by a transcutaneous blood gas monitor were obtained at the same sites on the dorsal foot and lower leg. Differences in Hb score and TcPO2 were analyzed by t-test for each measurement site and for the presence of peripheral arterial disease (PAD). The correlation between TcPO2 and Hb score was analyzed by Pearson's correlation coefficient. Results: Twenty-four limbs were tested. Hb score was higher (P < 0.001) and TcPO2 was lower (P = 0.056) in the dorsal foot compared to the lower leg. In the dorsal foot, Hb score was higher (P = 0.023) and TcPO2 was lower (P = 0.046) in patients with PAD compared to those without PAD. A significant negative correlation (r = -0.68; 95% confidence interval -0.85 to -0.38, P < 0.001) between TcPO2 and Hb score was observed in the dorsal foot. Conclusion: The negative correlation between TcPO2 and Hb score may reflect compensatory peripheral vasodilation due to occlusion or stenosis of central arteries. This study showed that Hb score measured by Antera 3D® may be related to skin blood flow.

Phoneme classification for speech synthesiser using differential EMG signals between muscles.

This paper proposes the use of differential electromyography (EMG) signals between muscles for phoneme classification, with which a Japanese speech synthesiser system can be constructed using fewer electrodes. In distinction from traditional methods using differential EMG signals between bipolar electrodes on the same muscle, an EMG signal is derived as differential between monopolar signals on two different muscles in the proposed method. Then, frequency-based feature patterns are extracted with filter banks, and classification of phonemes is realized by using a probabilistic neural network, which combines feature reduction and pattern classification processes in a single network structure. Experimental results show that the proposed method can achieve considerably high classification performance with fewer electrodes.