Chatbot dialogic reading boosts comprehension for Chinese kindergarteners with higher language skills.

Dialogic reading promotes early language and literacy development, but high-quality interactions may be inaccessible to disadvantaged children. This study examined whether a chatbot could deliver dialogic reading support comparable to a human partner for Chinese kindergarteners. Using a 2 × 2 factorial design, 148 children (83 girls; Mage = 70.07 months, SD = 7.64) from less resourced families in Beijing, China, were randomly assigned to one of four conditions: dialogic or non-dialogic reading techniques with either a chatbot or human partner. The chatbot provided comparable dialogic support to the human partner, enhancing story comprehension and word learning. Critically, the chatbot's effect on story comprehension was moderated by children's language proficiency rather than age or reading ability. This demonstrates that chatbots can facilitate dialogic reading and highlights the importance of considering children's language skills when implementing chatbot dialogic interventions.

Shape-margin knowledge augmented network for thyroid nodule segmentation and diagnosis.

BACKGROUND AND OBJECTIVE: Thyroid nodule segmentation is a crucial step in the diagnostic procedure of physicians and computer-aided diagnosis systems. However, prevailing studies often treat segmentation and diagnosis as independent tasks, overlooking the intrinsic relationship between these processes. The sequencial steps of these independent tasks in computer-aided diagnosis systems may lead to the accumulation of errors. Therefore, it is worth combining them as a whole by exploring the relationship between thyroid nodule segmentation and diagnosis. According to the diagnostic procedure of thyroid imaging reporting and data system (TI-RADS), the assessment of shape and margin characteristics is the prerequisite for radiologists to discriminate benign and malignant thyroid nodules. Inspired by TI-RADS, this study aims to integrate these tasks into a cohesive process, leveraging the insights from TI-RADS, thereby enhancing the accuracy and interpretability of thyroid nodule analysis. METHODS: Specifically, this paper proposes a shape-margin knowledge augmented network (SkaNet) for simultaneous thyroid nodule segmentation and diagnosis. Due to the visual feature similarities between segmentation and diagnosis, SkaNet shares visual features in the feature extraction stage and then utilizes a dual-branch architecture to perform thyroid nodule segmentation and diagnosis tasks respectively. In the shared feature extraction, the combination of convolutional feature maps and self-attention maps allows to exploitation of both local information and global patterns in thyroid nodule images. To enhance effective discriminative features, an exponential mixture module is introduced, combining convolutional feature maps and self-attention maps through exponential weighting. Then, SkaNet is jointly optimized by a knowledge augmented multi-task loss function with a constraint penalty term. The constraint penalty term embeds shape and margin characteristics through numerical computations, establishing a vital relationship between thyroid nodule diagnosis results and segmentation masks. RESULTS: We evaluate the proposed approach on a public thyroid ultrasound dataset (DDTI) and a locally collected thyroid ultrasound dataset. The experimental results reveal the value of our contributions and demonstrate that our approach can yield significant improvements compared with state-of-the-art counterparts. CONCLUSIONS: SkaNet highlights the potential of combining thyroid nodule segmentation and diagnosis with knowledge augmented learning into a unified framework, which captures the key shape and margin characteristics for discriminating benign and malignant thyroid nodules. Our findings suggest promising insights for advancing computer-aided diagnosis joint with segmentation.

Clinical analysis of anastomotic fistula after anal preservation in 358 cases of rectal cancer and construction of risk prediction model: a single-center retrospective study.

OBJECTIVE: To develop and validate a simple prediction model for postoperative anastomotic leakage (AL) in patients with rectal cancer who underwent Dixon surgery by combining preoperative and intraoperative risk factors. METHODS: We conducted a retrospective study on 358 patients who underwent Dixon surgery for rectal cancer in the Affiliated Hospital of Youjiang Medical University for Nationalities (Guangxi Zhuang Autonomous Region, China). Based on logistic regression, the prediction model of AL after Dixon surgery was established and verified. RESULTS: The incidence of postoperative AL in these patients was 9.2% (33/358). The results of logistic regression analysis showed that age ≥60 years, male, Tumor-Node-Metastasis (TNM) stage ≥IIIa, preoperative obstruction, and the distance from the tumor to the anus ≤7 cm were the risk factors for AL after Dixon surgery, and intraoperative defunctioning stoma was the protective factor for AL after rectal Dixon surgery (all P<0.05). The prediction model construction: Risk score =-4.275 + 0.851 × age + 1.047 × sex + 0.851 × distance + 0.934 × stage + 0.983 × obstruction. The area under receiver operating characteristic curve (ROC-AUC) was 0.762 (95% CI: 0.667-0.856). The best cutoff, sensitivity and specificity were 0.14, 79.60%, and 83.10%, respectively. Hosmer-Lemeshow: X2=6.876, P=0.550. Clinical validation results: the sensitivity, specificity, and accuracy of the model were 82.05%, 80.06%, and 80.25%, respectively. CONCLUSIONS: Both preoperative and intraoperative risk factors were used in the prognostic model. The prediction model established on this basis was well differentiated and highly calibrated, providing a good reference for the clinical prediction model of postoperative AL in rectal cancer patients undergoing Dixon surgery.