Machine learning based on clinical information and integrated CT radiomics to predict local recurrence of stage Ia lung adenocarcinoma after microwave ablation.

PURPOSE: To develop and compare three different machine learning-based models of clinical information and integrated radiomics features predicting the local recurrence of stage Ia lung adenocarcinoma after microwave ablation (MWA) for assisting clinical decision-making. MATERIALS AND METHODS: The data of 360 patients with stage Ia lung adenocarcinoma receiving MWA were included in the training (n = 208), internal test (n = 90), and external test (n = 64) sets based on the inclusion and exclusion criteria. The predictors associated with local recurrence were identified using univariate and multivariate analyses of clinical information. The integrated radiomics features were extracted from pre-MWA and post-MWA (scanned immediately after the ablation) CT images, and ten radiomics features were selected by the t-test and least absolute shrinkage and selection operator (LASSO). The L2-logistic regression of machine learning was applied for the clinical model, CT radiomics model and combined model including clinical predictors and radiomics features. Model performance was evaluated by the receiver operating characteristic (ROC) and decision curve analysis (DCA). RESULTS: The ablative margin was an independent clinical predictor (p = 0.001, odds ratio [OR] = 0.46, 95%CI: 0.29, 0.73). The combined model showed the highest area under the curve (AUC) value among the three models (training: 0.86, 95%CI: 0.81, 0.91; internal test: 0.93, 95%CI: 0.87, 0.98; external test: 0.89, 95%CI: 0.79, 0.96). CONCLUSION: The combined model could accurately predict the local recurrence of stage Ia lung adenocarcinoma after MWA to better support a clinical decision.

An electropermanent magnet valve for the onboard control of multi-degree of freedom pneumatic soft robots.

To achieve coordinated functions, fluidic soft robots typically rely on multiple input lines for the independent inflation and deflation of each actuator. Fluidic actuators are controlled by rigid electronic pneumatic valves, restricting the mobility and compliance of the soft robot. Recent developments in soft valve designs have shown the potential to achieve a more integrated robotic system, but are limited by high energy consumption and slow response time. In this work, we present an electropermanent magnet (EPM) valve for electronic control of pneumatic soft actuators that is activated through microsecond electronic pulses. The valve incorporates a thin channel made from thermoplastic films. The proposed valve (3 × 3 × 0.8 cm, 2.9 g) can block pressure up to 146 kPa and negative pressures up to -100 kPa with a response time of less than 1 s. Using the EPM valves, we demonstrate the ability to switch between multiple operation sequences in real time through the control of a six-DoF robot capable of grasping and hopping with a single pressure input. Our proposed onboard control strategy simplifies the operation of multi-pressure systems, enabling the development of dynamically programmable soft fluid-driven robots that are versatile in responding to different tasks.

Association between remote resistance exercises programs delivered by a smartphone application and skeletal muscle mass among elderly patients with type 2 diabetes- a retrospective real-world study.

Objective: We aimed to explore the relationship between remote resistance exercise programs delivered via a smartphone application and skeletal muscle mass among elderly patients with type 2 diabetes, utilizing real-world data. Methods: The resistance exercises were provided through Joymotion®, a web-based telerehabilitation smartphone application (Shanghai Medmotion Medical Management Co., Ltd). The primary outcome was the changes in skeletal muscle index (SMI) before and after the remote resistance exercises programs. The secondary outcomes were changes in skeletal muscle cross-sectional area (SMA), skeletal muscle radiodensity (SMD) and intermuscular adipose tissue (IMAT). Results: A total of 101 elderly patients with type 2 diabetes were analyzed. The participants had an average age of 72.9 ± 6.11 years for males and 74.4 ± 4.39 years for females. The pre- and post-intervention SMI mean (± SE) was 31.64 ± 4.14 vs. 33.25 ± 4.22 cm2/m2 in male, and 22.72 ± 3.24 vs. 24.28 ± 3.60 cm2/m2 in female respectively (all P < 0.001). Similarly, a statistically significant improvement in SMA, IMAT, and SMD for both male and female groups were also observed respectively (P < 0.001). Multiple linear regression models showed potential confounding factors of baseline hemoglobin A1c and duration of diabetes with changes in SMI in male, while hemoglobin A1c and high density lipoprotein cholesterol with changes in SMI in female. Conclusion: Remote resistance exercises programs delivered by a smartphone application were feasible and effective in helping elderly patients with type 2 diabetes to improve their skeletal muscle mass.

Factors that contribute to false-negative results in CT-guided transthoracic lung core-needle biopsy.

Objective: To investigate the possibility of false-negative occurrence of non-specific benign pathological results on CT-guided transthoracic lung core-needle biopsy and identify risk factors for false-negative results. Methods: The clinical, imaging, and surgical data of 403 lung biopsy patients were retrospectively analyzed. Patients were divided into true-negative and false-negative (FN) groups according to the final diagnosis. Univariate analysis was used to compare the variables in two groups for statistical differences, and multivariate analysis was used to clarify the risk factors associated with FN results. Results: Of the 403 lesions, 332 were finally confirmed as benign and 71 to be malignant, with a FN rate of 17.6%. Older patient age (P = 0.01), burr sign (P = 0.00), and pleural traction sign (P = 0.02) were independent risk factors for FN results. The area under the receiver operating characteristic (ROC) curve's area under curve (AUC) was 0.73. Conclusion: CT-guided transthoracic lung core-needle biopsy has a high diagnostic accuracy and low rate of FN results. Older patient age, the burr sign, and the pleural traction sign are independent risk factors for FN results that should be monitored prior to surgery to reduce the risk of FN results.

Functional characterization of a novel GUCA1A missense mutation (D144G) in autosomal dominant cone dystrophy: A novel pathogenic GUCA1A variant in COD.

Purpose: To elucidate the clinical phenotypes and pathogenesis of a novel missense mutation in guanylate cyclase activator A1A (GUCA1A) associated with autosomal dominant cone dystrophy (adCOD). Methods: The members of a family with adCOD were clinically evaluated. Relevant genes were captured before being sequenced with targeted next-generation sequencing and confirmed with Sanger sequencing. Sequence analysis was made of the conservativeness of mutant residues. An enzyme-linked immunosorbent assay (ELISA) was implemented to detect the cyclic guanosine monophosphate (cGMP) concentration. Then limited protein hydrolysis and an electrophoresis shift were used to assess possible changes in the structure. Coimmunoprecipitation was employed to analyze the interaction between GCAP1 and retGC1. Immunofluorescence staining was performed to observe the colocalization of GCAP1 and retGC1 in human embryonic kidney (HEK)-293 cells. Results: A pathogenic mutation in GUCA1A (c.431A>G, p.D144G, exon 5) was revealed in four generations of a family with adCOD. GUCA1A encodes guanylate cyclase activating protein 1 (GCAP1). D144, located in the EF4 loop involving calcium binding, was highly conserved in the species. GCAP1-D144G was more susceptible to hydrolysis, and the mobility of the D144G band became slower in the presence of Ca2+. At high Ca2+ concentrations, GCAP1-D144G stimulated retGC1 in the HEK-293 membrane to significantly increase intracellular cGMP protein concentrations. Compared with wild-type (WT) GCAP1, GCAP1-D144G had an increased interaction with retGC1, as detected in the coimmunoprecipitation assay. Conclusions: The newly discovered missense mutation in GUCA1A (p.D144G) might lead to an imbalance of Ca2+ and cGMP homeostasis and eventually, cause a significant variation in adCOD.