Transanal Minimally Invasive Surgery Versus Endoscopic Submucosal Dissection for Rectal Lesions: A Community Hospital Experience.

Background: To compare tumor margins and surgical outcomes between transanal minimally invasive surgery (TAMIS) and endoscopic submucosal dissection (ESD) for large or malignant rectal adenomatous polyps. Methods: Single institution retrospective analysis of patients who underwent TAMIS or ESD surgery. Results: In total, 30 consecutive patients with similar demographics who underwent either TAMIS (n = 19) or ESD (n = 11) were included. The median (interquartile range, IQR) tumor distances from the anal verge for TAMIS and ESD were 5 cm (3.5-8) and 3 cm (2-4.25) (P = 0.016). Four in TAMIS and two in ESD occupied more than half of the circumference of the bowel lumen. Five (four in situ and one stage 1) in TAMIS and two (one in situ and one stage 1) in ESD were malignant. The median specimen length, width, and height were 3.2 cm, 2.6 cm, and 1.0 cm and 3.5 cm, 2.0 cm, and 0.3 cm for TAMIS and ESD, respectively. There were no statistically significant differences in tumor circumference, malignant ratios, or specimen sizes. Resection margins were involved in two of the ESD, while none of the TAMIS were involved (P = 0.041). The median (IQR) operative time was 72 (62-89) minutes and 120 (90-180) minutes for TAMIS and ESD (P = 0.005). The median (IQR) follow-up time was 3.3 (0.3-11.7) and 0.9 (0.3-15.4) months for TAMIS and ESD. There were no morbidities, no mortalities, or local recurrences among the two groups. Conclusions: Both TAMIS and ESD were found to be feasible and safe in community hospital practice. Operative time was shorter, and there were no involved margins in TAMIS (versus ESD).

Deep convolutional neural network with fusion strategy for skin cancer recognition: model development and validation.

We aimed to develop an accurate and efficient skin cancer classification system using deep-learning technology with a relatively small dataset of clinical images. We proposed a novel skin cancer classification method, SkinFLNet, which utilizes model fusion and lifelong learning technologies. The SkinFLNet's deep convolutional neural networks were trained using a dataset of 1215 clinical images of skin tumors diagnosed at Taichung and Taipei Veterans General Hospital between 2015 and 2020. The dataset comprised five categories: benign nevus, seborrheic keratosis, basal cell carcinoma, squamous cell carcinoma, and malignant melanoma. The SkinFLNet's performance was evaluated using 463 clinical images between January and December 2021. SkinFLNet achieved an overall classification accuracy of 85%, precision of 85%, recall of 82%, F-score of 82%, sensitivity of 82%, and specificity of 93%, outperforming other deep convolutional neural network models. We also compared SkinFLNet's performance with that of three board-certified dermatologists, and the average overall performance of SkinFLNet was comparable to, or even better than, the dermatologists. Our study presents an efficient skin cancer classification system utilizing model fusion and lifelong learning technologies that can be trained on a relatively small dataset. This system can potentially improve skin cancer screening accuracy in clinical practice.

Power enhancement of piezoelectric transformers by adding heat transfer equipment.

It is known that piezoelectric transformers have several inherent advantages compared with conventional electromagnetic transformers. However, the maximum power capacity of piezoelectric transformers is not as large as electromagnetic transformers in practice, especially in the case of high output current. The theoretical power density of piezoelectric transformers calculated by stress boundary can reach 330 W/cm(3), but no piezoelectric transformer has ever reached such a high power density in practice. The power density of piezoelectric transformers is limited to 33 W/cm(3) in practical applications. The underlying reason is that the maximum passing current of the piezoelectric material (mechanical current) is limited by the temperature rise caused by heat generation. To increase this current and the power capacity, we proposed to add a thermal pad to the piezoelectric transformer to dissipate heat. The experimental results showed that the proposed techniques can increase by 3 times the output current of the piezoelectric transformer. A theoretical-phenomenological model which explains the relationship between vibration velocity and generated heat is also established to verify the experimental results.