Purpose: The molecular classification of breast cancer is crucial for effective treatment. The emergence of digital pathology has ushered in a new era in which weakly supervised learning leveraging whole-slide images has gained prominence in developing deep learning models because this approach alleviates the need for extensive manual annotation. Weakly supervised learning was employed to classify the molecular subtypes of breast cancer. Materials and Methods: Our approach capitalizes on two whole-slide image datasets: one consisting of breast cancer cases from the Korea University Guro Hospital (KG) and the other originating from The Cancer Genomic Atlas dataset (TCGA). Furthermore, we visualized the inferred results using an attention-based heat map and reviewed the histomorphological features of the most attentive patches. Results: The KG+TCGA-trained model achieved an area under the receiver operating characteristics value of 0.749. An inherent challenge lies in the imbalance among subtypes. Additionally, discrepancies between the two datasets resulted in different molecular subtype proportions. To mitigate this imbalance, we merged the two datasets, and the resulting model exhibited improved performance. The attentive patches correlated well with widely recognized histomorphologic features. The triple-negative subtype has a high incidence of high-grade nuclei, tumor necrosis, and intratumoral tumor-infiltrating lymphocytes. The luminal A subtype showed a high incidence of collagen fibers. Conclusion The artificial intelligence (AI) model based on weakly supervised learning showed promising performance. A review of the most attentive patches provided insights into the predictions of the AI model. AI models can become invaluable screening tools that reduce costs and workloads in practice.

Purpose: The molecular classification of breast cancer is crucial for effective treatment. The emergence of digital pathology has ushered in a new era in which weakly supervised learning leveraging whole-slide images has gained prominence in developing deep learning models because this approach alleviates the need for extensive manual annotation. Weakly supervised learning was employed to classify the molecular subtypes of breast cancer. Materials and Methods: Our approach capitalizes on two whole-slide image datasets: one consisting of breast cancer cases from the Korea University Guro Hospital (KG) and the other originating from The Cancer Genomic Atlas dataset (TCGA). Furthermore, we visualized the inferred results using an attention-based heat map and reviewed the histomorphological features of the most attentive patches. Results: The KG+TCGA-trained model achieved an area under the receiver operating characteristics value of 0.749. An inherent challenge lies in the imbalance among subtypes. Additionally, discrepancies between the two datasets resulted in different molecular subtype proportions. To mitigate this imbalance, we merged the two datasets, and the resulting model exhibited improved performance. The attentive patches correlated well with widely recognized histomorphologic features. The triple-negative subtype has a high incidence of high-grade nuclei, tumor necrosis, and intratumoral tumor-infiltrating lymphocytes. The luminal A subtype showed a high incidence of collagen fibers. Conclusion The artificial intelligence (AI) model based on weakly supervised learning showed promising performance. A review of the most attentive patches provided insights into the predictions of the AI model. AI models can become invaluable screening tools that reduce costs and workloads in practice.

Purpose: The molecular classification of breast cancer is crucial for effective treatment. The emergence of digital pathology has ushered in a new era in which weakly supervised learning leveraging whole-slide images has gained prominence in developing deep learning models because this approach alleviates the need for extensive manual annotation. Weakly supervised learning was employed to classify the molecular subtypes of breast cancer. Materials and Methods: Our approach capitalizes on two whole-slide image datasets: one consisting of breast cancer cases from the Korea University Guro Hospital (KG) and the other originating from The Cancer Genomic Atlas dataset (TCGA). Furthermore, we visualized the inferred results using an attention-based heat map and reviewed the histomorphological features of the most attentive patches. Results: The KG+TCGA-trained model achieved an area under the receiver operating characteristics value of 0.749. An inherent challenge lies in the imbalance among subtypes. Additionally, discrepancies between the two datasets resulted in different molecular subtype proportions. To mitigate this imbalance, we merged the two datasets, and the resulting model exhibited improved performance. The attentive patches correlated well with widely recognized histomorphologic features. The triple-negative subtype has a high incidence of high-grade nuclei, tumor necrosis, and intratumoral tumor-infiltrating lymphocytes. The luminal A subtype showed a high incidence of collagen fibers. Conclusion The artificial intelligence (AI) model based on weakly supervised learning showed promising performance. A review of the most attentive patches provided insights into the predictions of the AI model. AI models can become invaluable screening tools that reduce costs and workloads in practice.

Background/Aims: Pancreatic cancer poses significant challenges due to its tendency for late-stage diagnosis and high mortality rates. Cryoablation, a technique used to treat various types of cancer, has shown potential in enhancing the prognosis of pancreatic cancer when combined with other therapies. However, its implementation is often limited by the need for lengthy procedures and specialized equipment. This study aims to develop a cryoablation needle optimized for endoscopic ultrasonography to simplify its application in treating pancreatic cancer. Methods: The study involved conducting cryoablation experiments on swine liver tissue. It utilized cryo-needles to evaluate the extent of cell death across various temperatures and durations of cryoablation. Results: The cryoablation system, which employed liquid carbon dioxide, achieved rapid cooling, reaching temperatures below –60 °C within 30 seconds and maintained the cryoablation process for 200 seconds. These conditions resulted in necrosis of the liver tissue. Notable cellular changes were observed up to 15 mm away from the cryoablation needle. Conclusions This experimental study successfully demonstrated the efficacy of using a cryo-needle for cryoablation in swine liver tissue. Further trials involving pancreatic tissue are expected to verify its effectiveness, underscoring the importance of continued research to establish its role as a complementary therapy in pancreatic cancer treatment.

Chronic pancreatitis is a non-reversible, progressive inflammatory disease of the pancreas that is caused by repeated inflammation. It is characterized by morphological changes in the parenchyma and main pancreatic duct as pancreatic tissue is damaged and fibrosis progresses. These changes can lead to exocrine and endocrine dysfunction in patients with chronic pancreatitis, with major complications including pain, endocrine or exocrine insufficiency, and increased risk of pancreatic cancer. Of these, pancreatic exocrine insufficiency (PEI) is a state of malabsorption in which more than 15 g of fat is excreted per day. It can be diagnosed and assessed using several methods, but these methods are currently difficult to use in Korea and have low compliance, making them low utility. Recently, the PEI-questionnaire (PEI-Q) was developed to diagnose and assess patients’ exocrine insufficiency. It has been validated in foreign countries and has the potential to be used as a tool for diagnosing exocrine insufficiency with simple questions. This paper will review the diagnostic methods for existing exocrine insufficiency and the usefulness and utilization of the PEI-Q.

Background: Current international guidelines recommend against deep sedation as it is associated with worse outcomes in the intensive care unit (ICU). However, in Korea the prevalence of deep sedation and its impact on patients in the ICU are not well known. Methods: From April 2020 to July 2021, a multicenter, prospective, longitudinal, noninterventional cohort study was performed in 20 Korean ICUs. Sedation depth extent was divided into light and deep using a mean Richmond Agitation–Sedation Scale value within the first 48 hours. Propensity score matching was used to balance covariables; the outcomes were compared between the two groups. Results: Overall, 631 patients (418 [66.2%] and 213 [33.8%] in the deep and light sedation groups, respectively) were included. Mortality rates were 14.1% and 8.4% in the deep and light sedation groups (P = 0.039), respectively. Kaplan-Meier estimates showed that time to extubation (P < 0.001), ICU length of stay (P = 0.005), and death P = 0.041) differed between the groups. After adjusting for confounders, early deep sedation was only associated with delayed time to extubation (hazard ratio [HR], 0.66; 95% confidence inter val [CI], 0.55– 0.80; P < 0.001). In the matched cohort, deep sedation remained significantly associated with delayed time to extubation (HR, 0.68; 95% 0.56–0.83; P < 0.001) but was not associated with ICU length of stay (HR, 0.94; 95% CI, 0.79–1.13; P = 0.500) and in-hospital mortality (HR, 1.19; 95% CI, 0.65–2.17; P = 0.582). Conclusion In many Korean ICUs, early deep sedation was highly prevalent in mechanically ventilated patients and was associated with delayed extubation, but not prolonged ICU stay or in-hospital death.

Since the development of endoscopic ultrasound (EUS), the importance in the using EUS in pancreas and biliary tree has gradually increased. In order to properly observe the pancreas and biliary tree in EUS, it is necessary to learn landmark structures at each location, such as the stomach and duodenum, and to learn the positional relationship between them and normal structures. In this article, we will deal with the examination method at each position of the linear and radial EUS and reveal the normal EUS image.

Background: /Aim: Endoscopic retrograde cholangiopancreatography (ERCP) is one of the useful treatment for treating diseases that cause pancreatic duct stenosis. However, if the stenosis is severe, accessories other than the guide wire cannot pass through. This makes the ERCP procedure difficult. At this time, cystotome, a type of diathermic dilator seems to be an option. In this paper, we would like to discuss the experience of treatment procedures using cystotome. Methods: Patients who underwent ERCP at a single tertiary hospital from November 2019 to October 2022 were analyzed retrospectively. Among them, cystotome was used in eight procedures in seven patients who passed the guide wire but failed to pass the stenosis. Results: Technical success was achieved in all eight times. All the balloon catheter smoothly passed through the stenosis after cystotome was performed. In the second ERCP, it was a functional success for the balloon catheter to pass through stenosis without further treatment. This was achieved in 6 out of 8 (75.0%). In the case of complications, only one of the total successful procedures developed mild pancreatitis, which led to additional hospitalization for about two days after the procedure, and improved after conservative treatment. Conclusions Based on this, cystotome seems to be a good option as an alternative treatment to severe stenosis of the pancreatic duct.

Objective: This study investigated the change in the number of patients with head trauma according to the trauma mechanism among severely injured patients transferred to the emergency department of Wonju Severance Christian Hospital before and during the coronavirus disease 2019 (COVID-19) pandemic. Methods: Medical records (sex, age, diagnosis, trauma mechanism, and injury severity score) of patients referred to the emergency room between January 2018 and December 2019 and January 2020 and December 2021 were retrospectively reviewed, verified, and compared. Results: Between 2020 and 2021, the number of patients with traumatic brain injury decreased by 251 (32%). No significant differences were observed in sex, age, or time of accident. From 2020 to 2021, among the trauma mechanisms, the number of cases involving rolling down slightly reduced compared with those involving other mechanisms.Furthermore, cerebral contusions among intracranial lesions significantly increased during the COVID-19 pandemic. Conclusion Partial restrictions on social activities owing to COVID-19 are ongoing. Further investigation of the clinical characteristics of trauma patients over a longer period is required.