• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1224-1232
• /
• 2022

Whole body bone scan is the most frequently performed nuclear medicine imaging to evaluate bone metastasis in cancer patients. We evaluated the performance of a VGG16-based transfer learning classifier for bone scan images in which metastatic bone lesion was present. A total of 1,000 bone scans in 1,000 cancer patients (500 patients with bone metastasis, 500 patients without bone metastasis) were evaluated. Bone scans were labeled with abnormal/normal for bone metastasis using medical reports and image review. Subsequently, gradient-weighted class activation maps (Grad-CAMs) were generated for explainable AI. The proposed model showed AUROC 0.96 and F1-Score 0.90, indicating that it outperforms to VGG16, ResNet50, Xception, DenseNet121 and InceptionV3. Grad-CAM visualized that the proposed model focuses on hot uptakes, which are indicating active bone lesions, for classification of whole body bone scan images with bone metastases.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.150-156
• /
• 2023

For a doctor, diagnosing a patient's heart disease is not easy. It takes the ability and experience with high flying hours to be able to accurately diagnose the type of patient's heart disease based on the existing factors in the patient. Several studies have been carried out to develop tools to identify types of heart disease in patients. However, most only focus on the results of patient answers and lab results, the rest use only echocardiography data or electrocardiogram results. This research was conducted to test how accurate the results of the classification of heart disease by using two medical data, namely echocardiography and electrocardiogram. Three treatments were applied to the two medical data and analyzed using the decision tree approach. The first treatment was to build a classification model for types of heart disease based on echocardiography and electrocardiogram data, the second treatment only used echocardiography data and the third treatment only used electrocardiogram data. The results showed that the classification of types of heart disease in the first treatment had a higher level of accuracy than the second and third treatments. The accuracy level for the first, second and third treatment were 78.95%, 73.69% and 50%, respectively. This shows that in order to diagnose the type of patient's heart disease, it is advisable to look at the records of both the patient's medical data (echocardiography and electrocardiogram) to get an accurate level of diagnosis results that can be accounted for.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.4
• /
• pp.77-89
• /
• 2024

Knee osteoarthritis (OA) is a very common musculoskeletal disorder worldwide. The assessment of knee osteoarthritis, which requires a rapid and accurate initial diagnosis, is determined to be different depending on the currently dispersed classification system, and each classification system has different criteria. Also, because the medical staff directly sees and reads the X-ray pictures, it depends on the subjective opinion of the medical staff, and it takes time to establish an accurate diagnosis and a clear treatment plan. Therefore, in this study, we designed the stenosis length measurement algorithm and Osteophyte detection and length measurement algorithm, which are the criteria for determining the knee osteoarthritis grade, separately using CNN, which is a deep learning technique. In addition, we would like to create a grading system that integrates and complements the existing classification system and show results that match the judgments of actual medical staff. Based on publicly available OAI (Osteoarthritis Initiative) data, a total of 9,786 knee osteoarthritis data were used in this study, eventually achieving an Accuracy of 69.8% and an F1 score of 76.65%.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.6
• /
• pp.2527-2532
• /
• 2012

Objectives: The aim of the present study was to explore mechanisms underlying the effects of down-regulating ${\beta}$-catenin expression on esophageal carcinoma (EC) cells. Methods: Cell cycle distribution and apoptosis were determined using flow cytometry and annexin V apoptosis assay, respectively. Transmission electron microscopy (TEM) was used to examine changes in ultrastructure, while expression of cyclin D1 protein and mRNA was detected by western blot and real-time PCR. Proliferating cell nuclear antigen (PCNA) and extracellular signal-regulated kinase (ERK) 1-2 were evaluated by Western blot analysis. PCNA labeling index (LI) was determined by immunocytochemistry. Results: Compared with pGen-3-con transfected and Eca-109 cells, the percentage of G0/G1-phase pGen-3-CTNNB1 transfected cells was obviously increased (P<0.05), with no significant difference among the three groups with regard to apoptosis (P>0.05). pGen-3-CTNNB1 transfected cells exhibited obvious decrease in cyclin D1 mRNA and protein expression (P<0.05) and the ultrastructure of Eca-109 cells underwent a significant change after being transfected with pGen-3-CTNNB1, suggesting that down-regulating ${\beta}$-catenin expression can promote the differentiation and maturation. The expression of PCNA and the ERKI/2 phosphorylation state were also down-regulated in pGen-3-CTNNB1 transfected cells (P<0.05). At the same time, the PCNA labeling index was decreased accordingly (P<0.05). Conclusion: Inhibition of EC Eca-109 cellproliferation by down-regulating ${\beta}$-catenin expression could improve cell ultrastructure by mediating blockade in G0/G1 through inhibiting cyclin D1, PCNA and the MAPK pathway (p-ERK1/2).

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.1
• /
• pp.515-521
• /
• 2013

Purpose: To evaluate the efficacy and toxicity of induction chemotherapy followed by concurrent chemoradiotherapy (the treatment group) versus concurrent chemoradiotherapy with or without adjuvant chemotherapy (the control group) for locoregionally advanced nasopharyngeal carcinoma. Methods: The search strategy included Pubmed, Embase, the Cochrane Library, China National Knowledge Internet Web, Chinese Biomedical Database and Wanfang Database. We also searched reference lists of articles and the volumes of abstracts of scientific meetings. All randomized controlled trials were included for a meta-analysis performed with RevMan 5.1.0. The Grading of Recommendations Assessment, Development, and Evaluation system (GRADE) was used to rate the level of evidence. Results: Eleven studies were included. Risk ratios of 0.99 (95%CI 0.72-1.36), 0.37 (95%CI 0.20-0.69), 1.08 (95%CI 0.84-1.38), 0.98 (95%CI 0.75-1.27) were observed for 3 years overall survival, 3 years progression-free survival, 2 years loco-regional failure-free survival and 2 years distant metastasis failure-free survival. There were no treatment-related deaths in either group in the 11 studies. Risk ratios of 1.90 (95%CI 1.24-2.92), 2.67 (95%CI 0.64-11.1), 1.04 (95%CI 0.79-1.37), 0.98 (95%CI 0.27-3.52) were found for grade 3-4 leukopenia, grade 3-4 thrombocytopenia, grade 3-4 mucous membrane, and grade 3-4 hepatic hematologic and gastrointestinal toxicity, the most significant toxicities for patients. Conclusion: Compared with the control group, induction chemotherapy followed by concurrent chemoradiotherapy was well tolerated but could not significantly improve prognosis in terms of overall survival, loco-regional failure-free survival or distant metastasis failure-free survival.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.13
• /
• pp.5437-5442
• /
• 2014

Esophageal squamous cell carcinoma (ESCC) is one of the most malignancies with a poor prognosis. The phospholipase $C{\varepsilon}$ gene (PLCE1) encodes a novel ras-related protein effector mediating the effects of R-Ras on the actin cytoskeleton and membrane protrusion. However, molecular mechanisms pertinent to ESCC are unclear. We therefore designed PLCE1-special small interfering RNA and transfected to esophageal squamous cell (EC) 9706 cells to investigat the effects of PLCE1 gene silencing on the cell cycle and apoptosis of ESCC and indicate its important role in the development of ESCC. Esophageal cancer tissue specimens and normal esophageal mucosa were obtained and assayed by immunohistochemical staining to confirm overexpression of PLCE1 in neoplasias. Fluorescence microscopy was used to examine transfection efficiency, while the result of PLCE1 silencing was examined by reverse transcription (RT-PCR). Flow cytometry and annexin V apoptosis assays were used to assess the cell cycle and apoptosis, respectively. Expression of cyclin D1 and caspase-3 was detected by Western-blotting. The level of PLCE1 protein in esophageal cancer tissue was significantly higher than that in normal tissue. After transfection, the expression of PLCE1 mRNA in EC 9706 was significantly reduced, compared with the control group. Furthermore, flow cytometry results suggested that the PLCE1 gene silencing arrested the cell cycle in the G0/G1 phase; apoptosis was significantly higher than in the negative control group and mock group. PLCE1 gene silencing by RNAi resulted in decreased expression of cyclin D1 and increased expression of caspase-3. Our study suggests that PLCE1 may be an oncogene and play an important role in esophageal carcinogenesis through regulating proteins which control cell cycling and apoptosis.

• Advances in pediatric surgery
• /
• v.9 no.1
• /
• pp.6-11
• /
• 2003

This study reviews 14 years' experience treating esophageal atresia with special emphasis on the clinical profile and outcome. From May 1989 to February 2003, 65 cases of esophageal atresia (EA) were treated at Asan Medical Center. Boys outnumbered girls 2.4 to 1. Prematutity and low birth weight were 27.7% and 38.5%. Esophageal atresia with distal tracheoesophageal fistula (TEF) was the most common type (87.7%), followed by pure EA and H type fistula. Forty-six patients (70.8%) had one or more associated anomalies, cardiac malformations were the most common. Duodenal atresia was found in 7 cases. There were 6 patients (9.2 %) with VATER cluster. VACTERL cluster was present in 18 patients (27.7%), one of who fulfilled the complete syndrome. Waterston group A, B and C made up 21.5%, 40.0% and 38.5% of the total group. Surgical treatment was attempted in 63 patients and deferred in 2 who had severe associated malformations. For EA with distal TEF, primary esophago esophagostomy was carried out in 51 cases, and division of TEF and gastrostomy in 4 cases and no operation in 2 cases. For pure EA, colonic graft was done in 2 after gastrostomy and esophagostomy, and esophago esophagostomy was performed in 2 after gastrostomy. Two patients with pure EA are waiting for the second operation after gastrostomy. Division of TEF was carried out in 2 cases with H type TEF. The overall survival rate was 76.9%, and survival by Waterston classification was 100% in group A, 80.8% in B and 60.0% in C. Thorough workup for associated anomalies, interdepartmental approach and more careful surgical decision and technique are required to improve the outcome of EA.

• Advances in pediatric surgery
• /
• v.14 no.1
• /
• pp.37-47
• /
• 2008

Recently, the survival rates of extremely low-birth-weight (ELBW) infants have improved with the development of neonatal intensive care. However, these infants were susceptible to intestinal perforation due to prematurity, fluid restriction, and injection of indomethacin, etc. Because of the risks of transportation, anesthesia and surgery itself, peritoneal drainage has been compared with laparotomy. Through our experience, we investigate the usefulness of peritoneal drainage retrospectively. From 1997 to 2007, six ELBW (M:F=5:1) underwent primary peritoneal drainage for intestinal perforation. Their median birth weight was 685g (405~870) and gestational age was $25^{+1}$ weeks ($24^{+3}{\sim}27^{+0}$). We noticed the intestinal perforation at median 10.5 days (8~18) after birth, and placed Penrose drain or Jackson-Pratt drain through right lower quadrant incision under local anesthesia. The cause of intestinal perforation was necrotizing enterocolitis in one patient, but that of the others was not clear. Three patients who showed normal platelet count and stable vital signs recovered uneventfully. Two patients (birth weight less than 500g) who showed unstable vital signs and low platelet count (12,000 / $mm^3$ to 30,000 / $mm^3$)expired despite aggressive resuscitation. One patient required laparotomy due to persistent intestinal obstruction after drain removal and survived. Our experience shows that peritoneal drainage was an acceptable treatment for ELBW infants and the prognosis was related to vital sign and platelet count at the time of intestinal perforation, and birth weight.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.137-142
• /
• 2024

Recently, deep learning technology has become those methods as de facto standards in the area of medical data representation. But, deep learning inherently requires a large amount of training data, which poses a challenge for its direct application in the medical field where acquiring large-scale data is not straightforward. Additionally, brain signal modalities also suffer from these problems owing to the high variability. Research has focused on designing deep neural network structures capable of effectively extracting spectro-spatio-temporal characteristics of brain signals, or employing self-supervised learning methods to pre-learn the neurophysiological features of brain signals. This paper analyzes methodologies used to handle small-scale data in emerging fields such as brain-computer interfaces and brain signal-based state prediction, presenting future directions for these technologies. At first, this paper examines deep neural network structures for representing brain signals, then analyzes self-supervised learning methodologies aimed at efficiently learning the characteristics of brain signals. Finally, the paper discusses key insights and future directions for deep learning-based brain signal analysis.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.7
• /
• pp.4121-4126
• /
• 2013

Objective: Low dose radiation may stimulate the growth and development of animals, increase life span, enhance fertility, and downgrade the incidence of tumor occurrence.The aim of this study was to investigate the antitumor effect and hormesis in an erythrocyte system induced by low-dose radiation. Methods: Kunming strain male mice were subcutaneously implanted with S180 sarcoma cells in the right inguen as an experimental in situ animal model. Six hours before implantation, the mice were given 75mGy whole body X-ray radiation. Tumor growth was observed 5 days later, and the tumor volume was calculated every other day. Fifteen days later, all mice were killed to measure the tumor weight, and to observe necrotic areas and tumor-infiltration-lymphoreticular cells (TILs). At the same time, erythrocyte immune function and the level of 2,3-diphosphoglyceric acid (2,3-DPG) were determined. Immunohistochemical staining was used to detect the expression of EPO and VEGFR of tumor tissues. Results: The mice pre-exposed to low dose radiation had a lower tumor formation rate than those without low dose radiation (P < 0.05). The tumor growth slowed down significantly in mice pre-exposed to low dose radiation; the average tumor weight in mice pre-exposed to low dose radiation was lighter too (P < 0.05). The tumor necrosis areas were larger and TILs were more in the radiation group than those of the group without radiation. The erythrocyte immune function, the level of 2,3-DPG in the low dose radiation group were higher than those of the group without radiation (P < 0.05). After irradiation the expression of EPO of tumor tissues in LDR group decreased with time. LDR-24h, LDR-48h and LDR-72h groups were all statistically significantly different from sham-irradiation group. The expression of VEGFR also decreased, and LDR-24h group was the lowest (P < 0.05). Conclusion: Low dose radiation could markedly increase the anti-tumor ability of the organism and improve the erythrocyte immune function and the ability of carrying $O_2$. Low-dose total body irradiation, within a certain period of time, can decrease the expression of hypoxia factor EPO and VEGFR, which may improve the situation of tumor hypoxia and radiosensitivity of tumor itself.