Deep Learning Analysis for Predicting Tumor Spread through Air Space in Early-Stage Lung Adenocarcinoma Pathology Images.

The presence of spread through air spaces (STASs) in early-stage lung adenocarcinoma is a significant prognostic factor associated with disease recurrence and poor outcomes. Although current STAS detection methods rely on pathological examinations, the advent of artificial intelligence (AI) offers opportunities for automated histopathological image analysis. This study developed a deep learning (DL) model for STAS prediction and investigated the correlation between the prediction results and patient outcomes. To develop the DL-based STAS prediction model, 1053 digital pathology whole-slide images (WSIs) from the competition dataset were enrolled in the training set, and 227 WSIs from the National Taiwan University Hospital were enrolled for external validation. A YOLOv5-based framework comprising preprocessing, candidate detection, false-positive reduction, and patient-based prediction was proposed for STAS prediction. The model achieved an area under the curve (AUC) of 0.83 in predicting STAS presence, with 72% accuracy, 81% sensitivity, and 63% specificity. Additionally, the DL model demonstrated a prognostic value in disease-free survival compared to that of pathological evaluation. These findings suggest that DL-based STAS prediction could serve as an adjunctive screening tool and facilitate clinical decision-making in patients with early-stage lung adenocarcinoma.

In situ HER2 RNA expression as a predictor of pathologic complete response of HER2-positive breast cancer patients receiving neoadjuvant chemotherapy and anti-HER2 targeted treatment.

BACKGROUND: Immunohistochemistry (IHC) and in situ hybridization (ISH) remain standard biomarkers for therapeutic decisions in human epidermal growth factor 2 (HER2)-positive breast cancers (BCs); however, they are insufficient to explain the heterogeneous anti-HER2 response. METHODS: We aimed to investigate the correlation of in situ HER2 RNA expression (isHRE), using RNAscope, with HER2 biomarkers and the impact of isHRE on the pathological complete response (pCR) rates of 278 patients with HER2 IHC/fluorescence ISH (FISH)-positive BC receiving neoadjuvant chemotherapy and anti-HER2 targeted treatment (NCTT). RESULTS: We validated HER2 RNAscope scoring as a semiquantitative method to determine isHRE and showed a positive correlation between RNAscope scores and pCR rates, with particularly different rates between patients with a score of 5 versus 1-4 BCs (66.7% vs. 15.9%, p < 0.0001). There were higher RNAscope scores and pCR rates in patients with HER2 IHC 3 + versus IHC 2+/FISH + BCs and HER2 RNAscope scores and pCR rates showed similar non-linear positive correlations with HER2 copy numbers and HER2/centromere 17 ratios. Moreover, in each HER2-positive IHC/FISH category, higher pCR rates were observed in patients with RNAscope scores of 5 versus 1-4 BC. Patients achieving pCR had BCs with notably higher HER2 RNAscope scores. Multivariate analysis identified HER2 RNAscope 5 as a strong pCR predictor [odds ratio = 10.865, p < 0.001]. The combined impact of multivariate analysis-defined pCR predictors demonstrated that a higher pCR rate was observed in patients with a score of 5 versus a score of 1-4 BCs regardless of the status of hormone receptor and mono-or dual anti-HER2 blockade. CONCUSIONS: Our results demonstrated that high isHRE (RNAscope score 5) is a strong pCR predictor in patients with HER2-positive BCs receiving NCTT, highlighting the complementary role of isHRE in stratifying HER2 status in tissue. Such stratification is relevant to anti-HER2 therapeutic efficacy, particularly using the cutoff of score 1-4 versus 5.

Evaluating the performance of an AI-powered VBAC prediction system within a decision-aid birth choice platform for shared decision-making.

Background: Vaginal birth after cesarean (VBAC) is generally regarded as a safe and viable birthing option for most women with prior cesarean delivery. Nonetheless, concerns about heightened risks of adverse maternal and perinatal outcomes have often dissuaded women from considering VBAC. This study aimed to assess the performance of an artificial intelligence (AI)-powered VBAC prediction system integrated into a decision-aid birth choice platform for shared decision-making (SDM). Materials and Methods: Employing a retrospective design, we collected medical records from a regional hospital in northern Taiwan from January 2019 to May 2023. To explore a suitable model for tabular data, we compared two prevailing modeling approaches: tree-based models and logistic regression models. We subjected the tree-based algorithm, CatBoost, to binary classification. Results: Forty pregnant women with 347 records were included. The CatBoost model demonstrated a robust performance, boasting an accuracy rate of 0.91 (95% confidence interval (CI): 0.86-0.94) and an area under the curve of 0.89 (95% CI: 0.86-0.93), surpassing both regression models and other boosting techniques. CatBoost captured the data characteristics on the significant impact of gravidity and the positive influence of previous vaginal birth, reinforcing established clinical guidelines, as substantiated by the SHapley Additive exPlanations analysis. Conclusion: Using AI techniques offers a more accurate assessment of VBAC risks, boosting women's confidence in selecting VBAC as a viable birthing option. The seamless integration of AI prediction systems with SDM platforms holds a promising potential for enhancing the effectiveness of clinical applications in the domain of women's healthcare.

A case of primary pleural epithelioid hemangioendothelioma achieving stable disease with paclitaxel treatment: A case report and literature review.

Epithelioid hemangioendothelioma (EHE) is a rare vascular neoplasm with a clinical behaviour that falls between a benign hemangioma and a high-grade angiosarcoma. Pleural EHE is exceptionally rare, and its prognosis is grim, with most patients experiencing survival of less than 1 year. Here, we present a case of pleural EHE in a 45-year-old woman with a month-long history of right-sided pleuritic chest pain. Chest computed tomography revealed consolidation, atelectasis of the right lung, right pleural thickening, and pleural effusion. She underwent video-assisted thoracoscopic surgery for decortication and was diagnosed with conclusively pleural EHE, showing a CAMTA1 rearrangement. Paclitaxel treatment, administered once weekly on days 1, 8 and 15 of a 28-day cycle, resulted in a stable disease after 12 cycles. Managing patients with pleural EHE is challenging because there are still no established standard treatments. Our case achieved 11-month progression-free survival following paclitaxel treatment.

Continuous Myocardial Perfusion during Distal Anastomosis of Acute Type A Aortic Dissection.

BACKGROUND: The effect of continuous myocardial perfusion (CMP) on the surgical results of acute type A aortic dissection (ATAAD) remains unclear. METHODS: From January 2017 to March 2022, 141 patients who underwent ATAAD (90.8%) or intramural hematoma (9.2%) surgery were reviewed. Fifty-one patients (36.2%) received proximal-first aortic reconstruction and CMP during distal anastomosis. Ninety patients (63.8%) underwent distal-first aortic reconstruction and were placed in traditional cold blood cardioplegic arrest (CA; 4°C, 4:1 blood-to-Plegisol) throughout the procedure. The preoperative presentations and intraoperative details were balanced using inverse probability of treatment weighting (IPTW). Their postoperative morbidity and mortality were analyzed. RESULTS: The median age was 60 years. The incidence of arch reconstruction in the unweighted data was higher in the CMP compared with the CA group (74.5 vs 52.2%, p = 0.017) but was balanced after IPTW (62.4 vs 58.9%, p = 0.932, standardized mean difference = 0.073). The median cardiac ischemic time was lower in the CMP group (60.0 vs 130.9 minutes, p < 0.001), but cerebral perfusion time and cardiopulmonary bypass time were similar. The CMP group did not demonstrate any benefit in the reduction of the postoperative maximum creatine kinase-MB ratio (4.4 vs 5.1% in CA, p = 0.437) or postoperative low cardiac output (36.6 vs 24.8%, p = 0.237). Surgical mortality was comparable between groups (15.5% in CMP vs 7.5% in the CA group, p = 0.265). CONCLUSION: Application of CMP during distal anastomosis in ATAAD surgery, irrespective of the extent of aortic reconstruction, reduced myocardial ischemic time but did not improve cardiac outcome or mortality.

Deterministic nuclear reprogramming of mammalian nuclei to a totipotency-like state by Amphibian meiotic oocytes for stem cell therapy in humans.

The ultimate aim of nuclear reprogramming is to provide stem cells or differentiated cells from unrelated cell types as a cell source for regenerative medicine. A popular route towards this is transcription factor induction, and an alternative way is an original procedure of transplanting a single somatic cell nucleus to an unfertilized egg. A third route is to transplant hundreds of cell nuclei into the germinal vesicle (GV) of a non-dividing Amphibian meiotic oocyte, which leads to the activation of silent genes in 24 h and robustly induces a totipotency-like state in almost all transplanted cells. We apply this third route for potential therapeutic use and describe a procedure by which the differentiated states of cells can be reversed so that totipotency and pluripotency gene expression are regained. Differentiated cells are exposed to GV extracts and are reprogrammed to form embryoid bodies, which shows the maintenance of stemness and could be induced to follow new directions of differentiation. We conclude that much of the reprogramming effect of eggs is already present in meiotic oocytes and does not require cell division or selection of dividing cells. Reprogrammed cells by oocytes could serve as replacements for defective adult cells in humans.

An innovative web-based decision-aid about birth after cesarean for shared decision making in Taiwan: study protocol for a randomized control trial.

BACKGROUND: Taiwan has a high national caesarean rate coupled with a low vaginal birth after caesarean (VBAC) rate. This study aims to develop and evaluate a web-based decision-aid with communication support tools, to increase shared decision making (SDM) about birth after caesarean. METHODS: A quantitative approach will be adopted using a randomized pre-test and post-test experimental design in a medical centre in northern Taiwan. The web-based decision aid consists of five sections. Section 1 provides a two-part video to introduce SDM and how to participate in SDM. Section 2 presents an overview of functions and features of the birth decision-aid. Section 3 presents relevant VBAC information, including definitions, benefits and risks, and an artificial intelligence (AI) calculator for rate and likelihood of VBAC success. Section 4 presents the information regarding elective repeat caesarean delivery (ERCD), involving definitions, benefits, and risks. Section 5 comprises four steps of decision making to meet women's values and preferences. Pregnant women who have had one previous caesarean and are eligible for VBAC, will be recruited at 14-16 weeks. Participants will complete a baseline survey prior to random allocation to either the control group (usual care) or intervention group (usual care plus an AI-decision aid). A follow up survey at 35-38 weeks will measure change in decisional conflict, knowledge, birth mode preference, and decision-aid acceptability. Actual birth outcomes and satisfaction will be assessed one month after birth. DISCUSSION: The innovative web-based decision-aid with support tools will help to promote pregnant women's decision-making engagement and communication with their providers and improve opportunities for supportive communication about VBAC SDM in Taiwan. Linking web-based AI data analysis into the medical record will also be assessed for feasibility during implementation in clinical practice. TRIAL REGISTRATION: ClinicalTrials.gov identifier (NCT05091944), Registered on October 24, 2021.

Real-Time Internet of Medical Things System for Detecting Blood Leakage during Hemodialysis Using a Novel Multiple Concentric Ring Sensor.

Venous needle dislodgement (VND) is a major healthcare safety concern in patients undergoing hemodialysis. Although VND is uncommon, it can be life-threatening. The main objective of this study was to implement a real-time multi-bed monitoring system for VND by combining a novel leakage-detection device and IoMT (Internet of Medical Things) technology. The core of the system, the Acusense IoMT platform, consisted of a novel leakage-detection patch comprised of multiple concentric rings to detect blood leakage and quantify the leaked volume. The performance of the leakage-detection system was evaluated on a prosthetic arm and clinical study. Patients with a high risk of blood leakage were recruited as candidates. The system was set up in a hospital, and the subjects were monitored for 2 months. During the pre-clinical simulation experiment, the system could detect blood leakage volumes from 0.3 to 0.9 mL. During the test of the IoMT system, the overall success rate of tests was 100%, with no lost data packets. A total of 701 dialysis sessions were analyzed, and the accuracy and sensitivity were 99.7% and 90.9%, respectively. Evaluation questionnaires showed that the use of the system after training changed attitudes and reduced worry of the nursing staff. Our results show the feasibility of using a novel detector combined with an IoMT system to automatically monitor multi-bed blood leakage. The innovative concentric-circle design could more precisely control the warning blood-leakage threshold in any direction to achieve clinical cost-effectiveness. The system reduced the load on medical staff and improved patient safety. In the future, it could also be applied to home hemodialysis for telemedicine during the era of COVID-19.

A Comparison of the Use of Two-Dimensional and Three-Dimensional Visualization Systems by Novice and Experienced Microsurgeons in Microsurgical Vessel Anastomosis: An Analysis Using the Chicken Model.

BACKGROUND: Various studies have discussed the benefits of applying three-dimensional (3D) techniques, specifically its advantages with respect to ergonomics, feasibility, and the rate of learning achievable in microsurgery training. However, no study has been conducted that compares the operator experience of using two-dimensional (2D) and 3D systems in microsurgical training. The aim of this study is to compare 2D- and 3D-assisted microsurgical training in novices based on anastomosis of chicken femoral arteries. METHODS: The participants were grouped by previous microsurgical experience. Group A includes novice participants. Group B includes 2D-experienced participants. Group C includes both participants in groups A and B. A questionnaire composed of 10 parameters in the field of image quality, dexterity, ergonomic, and feasibility will be filled out after each participant finished their anastomoses by the 2D and 3D systems. RESULTS: The results demonstrated 3D system was scored better on "field of view" (p = 0.004), "less tremor" (p = 0.005), "neck/upper back comfort" (p = 0.043), "lower back comfort" (p = 0.015), "technical feasibility" (p = 0.020), and "educational feasibility" (p = 0.004) in group A (N = 12). In group B (N = 9), 3D system was scored better on "field of view" (p = 0.041) but worse on "image resolution" (p = 0.031). CONCLUSION: With the 3D visualization system for microsurgical anastomosis of chicken femoral model, there are significant improvements in the field of view, stability, ergonomics, and educational value compared with 2D system among all participants. Accordingly, 3D-assisted microsurgery training can be a novel and potential popular training method.

Claudin-18 Immunohistochemical Staining Facilitates the Identification of Metastatic Carcinoma of Gastric or Pancreatic Origin in Effusion Specimens.

Determining the primary origin of a malignant effusion remains a common challenge for cytopathologists. Although immunohistochemical (IHC) markers are available for most primary sites, ideal IHC markers for metastatic gastric adenocarcinoma and pancreatic ductal adenocarcinoma are lacking, and related interpretation is often hindered by mesothelial cells. We recently revealed that claudin-18 IHC staining is useful for identifying the stomach and pancreas as the primary sites of metastatic adenocarcinoma. Thus, we assessed the use of claudin-18 IHC staining in 111 cell blocks obtained from various metastatic cancers and specimens negative for malignancy. Positive membranous claudin-18 staining was noted in all 10 (100%) metastatic pancreatic ductal adenocarcinomas, 9 (90%) of 10 gastric adenocarcinomas, and 1 (9%) of 11 nonmucinous lung adenocarcinomas. The cases of metastatic mucinous carcinomas of lung origin (1 case) and ovarian origin (1 case) were also positive for claudin-18. The other remaining 89 cases showed variable cytoplasmic staining on some cells (73 cases) or complete absence of staining (16 cases). After normalization to the tumor frequency, the sensitivity and specificity for identifying the stomach or pancreas as primary tumor sites in ascites were 95% (confidence interval: 0.83-0.99) and 99% (confidence interval: 0.94-1), respectively. In conclusion, membranous claudin-18 staining is a useful marker for metastatic gastric adenocarcinoma and pancreatic ductal adenocarcinoma in effusion specimens.