Technology and Health Care - Volume 33, issue 1

Authors: Muniasamy, Anandhavalli

Article Type: Research Article

Abstract: BACKGROUND: A daily activity routine is vital for overall health and well-being, supporting physical and mental fitness. Consistent physical activity is linked to a multitude of benefits for the body, mind, and emotions, playing a key role in raising a healthy lifestyle. The use of wearable devices has become essential in the realm of health and fitness, facilitating the monitoring of daily activities. While convolutional neural networks (CNN) have proven effective, challenges remain in quickly adapting to a variety of activities. OBJECTIVE: This study aimed to develop a model for precise recognition of human activities to …revolutionize health monitoring by integrating transformer models with multi-head attention for precise human activity recognition using wearable devices. METHODS: The Human Activity Recognition (HAR) algorithm uses deep learning to classify human activities using spectrogram data. It uses a pretrained convolution neural network (CNN) with a MobileNetV2 model to extract features, a dense residual transformer network (DRTN), and a multi-head multi-level attention architecture (MH-MLA) to capture time-related patterns. The model then blends information from both layers through an adaptive attention mechanism and uses a SoftMax function to provide classification probabilities for various human activities. RESULTS: The integrated approach, combining pretrained CNN with transformer models to create a thorough and effective system for recognizing human activities from spectrogram data, outperformed these methods in various datasets – HARTH, KU-HAR, and HuGaDB produced accuracies of 92.81%, 97.98%, and 95.32%, respectively. This suggests that the integration of diverse methodologies yields good results in capturing nuanced human activities across different activities. The comparison analysis showed that the integrated system consistently performs better for dynamic human activity recognition datasets. CONCLUSION: In conclusion, maintaining a routine of daily activities is crucial for overall health and well-being. Regular physical activity contributes substantially to a healthy lifestyle, benefiting both the body and the mind. The integration of wearable devices has simplified the monitoring of daily routines. This research introduces an innovative approach to human activity recognition, combining the CNN model with a dense residual transformer network (DRTN) with multi-head multi-level attention (MH-MLA) within the transformer architecture to enhance its capability. Show more

Keywords: Wearable sensors, human activity recognition, dense-residual connection, attention, CNN, transformers

DOI: 10.3233/THC-241064

Citation: Technology and Health Care, vol. 33, no. 1, pp. 395-409, 2025

Authors: Wang, Bing | Wu, Zehui | Liu, Gang | Liu, Ben | Yang, Wanchao | Yang, Chao | Shi, Lianghui

Article Type: Research Article

Abstract: BACKGROUND: Advancements in laparoscopic technology have popularized laparoscopic total gastrectomy over traditional open surgery, yet postoperative complications like anastomotic leakage and stenosis persist, particularly in esophagojejunostomy. To address this, since 2017, the authors have introduced the “Pant-Shaped” esophagojejunostomy as an improvement over the classic Roux-en-Y method, especially beneficial for patients with small intestinal diameters or those with gastric body cancer or Siewert III. OBJECTIVE: To assess the viability and safety of employing ‘Pant-Shaped’ anastomosis following laparoscopic-assisted total gastrectomy. METHODS: A method of descriptive case study was used. In our department of the First …Affiliated Hospital of Wannan Medical College, records of 210 patients who underwent laparoscopic-assisted total gastrectomy for gastric body cancer or adenocarcinoma at the junction of esophagus and jejunum with “Pant-Shaped” anastomosis between January 2017 and December 2022 were examined. Clinicopathological features and postoperative conditions were also examined and assessed. RESULTS: The mean age of the 164 male and 46 female research participants was 69.2 ± 8.3 years. There was a mean estimated blood loss of 63.4 ± 29.7 ml, an anastomosis time of 25.9 ± 3.0 minutes, an operation time of 208.2 ± 40.4 minutes, and a postoperative hospital stay of 12.2 ± 8.0 days. Nine patients (4.3%) experienced postoperative problems (Clavien-Dindo > grade II), including two episodes of anastomotic leakage that were resolved with irrigation and drainage, anti-infection therapy, and nutritional assistance. After an unforeseen reoperation, two cases of duodenal stump leaking were resolved. Anastamotic hemorrhage was treated with hemostasis and blood transfusion, and the patient made a full recovery. Due to a Peterson’s hernia, one patient required emergent open surgery. three months subsequent to LATG. CONCLUSIONS: The “Pant-Shaped” anastomosis method after laparoscopic-assisted total gastrectomy is simple, easy to promote, and has fewer complications. It is a safe and feasible modified method for esophagojejunostomy, especially suitable for patients with poor intestinal dilation and contraction ability and small jejunal diameter. Show more

Keywords: Laparoscopy, gastric cancer, gastrectomy, anastomosis, Roux-en-Y

DOI: 10.3233/THC-241093

Citation: Technology and Health Care, vol. 33, no. 1, pp. 411-418, 2025

Authors: Khaleel Sallam Ma’aitah, Mohammad | Helwan, Abdulkader

Article Type: Research Article

Abstract: BACKGROUND: Deep learning has demonstrated superior performance over traditional methods for the estimation of heart rates in controlled contexts. However, in less controlled scenarios this performance seems to vary based on the training dataset and the architecture of the deep learning models. OBJECTIVES: In this paper, we develop a deep learning-based model leveraging the power of 3D convolutional neural networks (3DCNN) to extract temporal and spatial features that lead to an accurate heart rates estimation from RGB no pre-defined region of interest (ROI) videos. METHODS: We propose a 3D DenseNet with a 3D …temporal transition layer for the estimation of heart rates from a large-scale dataset of videos that appear more hospital-like and real-life than other existing facial video-based datasets. RESULTS: Experimentally, our model was trained and tested on this less controlled dataset and showed heart rate estimation performance with root mean square error (RMSE) of 8.68 BPM and mean absolute error (MAE) of 3.34 BPM. CONCLUSION: Moreover, we show that such a model can also achieve better results than the state-of-the-art models when tested on the VIPL-HR public dataset. Show more

Keywords: Heart rate, deep learning, rPPG, 3D convolutional neural networks, no pre-defined region of interest (ROI)

DOI: 10.3233/THC-241104

Citation: Technology and Health Care, vol. 33, no. 1, pp. 419-430, 2025

Authors: Zhang, Xibing | Yang, Fuli | Han, Lei | Su, Qiuming | Gao, Yang | Wu, Ruichao | Wang, Duo | Li, Wang | Zheng, Kepu | Liu, Fang | Ran, Jianghua

Article Type: Research Article

Abstract: BACKGROUND: Liver fibrosis is a progressive liver disease with increasing incidence, yet its underlying pathogenic mechanisms remain incompletely understood. OBJECTIVE : This study aims to explore potential therapeutic targets for liver fibrosis using weighted gene co-expression network analysis (WGCNA) and experimental validation. METHODS: We retrieved the microarray data (GSE174099) from the GEO database and performed differential expression analysis and WGCNA to identify co-expression modules associated with liver fibrosis. A module with the highest correlation to liver fibrosis was selected for further analysis. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway …enrichment analyses were conducted to investigate the biological functions and signaling pathways of the identified genes. Protein-protein interaction (PPI) networks were constructed using the STRING database. The correlation between core genes and immune cells was analyzed with the CIBERSORT algorithm. Additionally, pathological and molecular biology experiments were performed to validate the expression levels of core genes in liver tissue, including HE and Masson staining, immunohistochemistry, RT-qPCR, and Western blotting. RESULTS: We identified a total of 86 intersecting genes from the differential expression analysis and WGCNA. GO enrichment analysis revealed that these genes were involved in processes such as cellular response to cAMP, collagen-containing extracellular matrix, and G protein-coupled receptor binding. KEGG pathway analysis highlighted the involvement of these genes in pathways like Cell Adhesion Molecules and the PI3K-Akt signaling pathway. Using Cytoscape software, we identified four core genes: Cftr, Cldn4, Map2, and Spp1. Pathological examinations showed that the experimental group exhibited significant fibrous tissue proliferation compared to the control group. Immunohistochemistry, RT-qPCR, and Western blotting analyses confirmed that these core genes were significantly upregulated in the experimental group (P < 0.05). CONCLUSION: This study identified four key genes (Cftr, Cldn4, Map2, Spp1) that are significantly associated with liver fibrosis. These genes are upregulated in liver fibrosis and could potentially as biomarkers for diagnosis and targets for therapeutic interventions. Show more

Keywords: Liver fibrosis, WGCNA, protein-protein interaction, signal transduction, gene expression

DOI: 10.3233/THC-241142

Citation: Technology and Health Care, vol. 33, no. 1, pp. 431-448, 2025

Authors: Han, Zhen | Liu, Yangan | Tan, Ming | Hua, Zhaolai | Dai, Chun

Article Type: Research Article

Abstract: BACKGROUND: Surgery remains the most effective treatment for colon cancer. However, there are still controversies regarding the tumor treatment effect, prognosis, and perioperative impact of complete mesocolic excision (CME) surgery in colon carcinoma resection. OBJECTIVE: This study aims to compare laparoscopic complete mesocolic excision (LCME) and traditional open D2 radical surgery in colon carcinoma resection through a retrospective analysis using 1:1 propensity score matching (PSM). METHODS: 98 cases undergoing LCME or open D2 colon carcinoma resection at our hospital from January 2014 to November 2021 were retrospectively collected. After excluding cases and 1:1 matching using …PSM based on baseline clinical data, 86 patients were assigned in research queue. 43 patients were in each group. Two groups were compared for general clinical baseline indicators. Surgical results and postoperative adverse events of patients were also compared. Disease-free survival (DFS) rate and overall survival (OS) rate was analyzed. RESULTS: After 1:1 PSM matching, there was no statistically significant differences in baseline data between the LCME group and D2 group (P > 0.05). LCME was characterized by longer total duration of surgery (P < 0.001), less intraoperative bleeding volume (P < 0.001), more postoperative drainage fluid volume (P < 0.001), greater number of lymph nodes retrieved (P = 0.014). No statistically differences was observed regarding intraoperative blood transfusion, hospital stay, Clavien-Dindo complicating disease classification (all P > 0.05), 1 and 3-year DFS rate (P = 0.84) and OS rate (P ⩾ 0.1). CONCLUSION: LCME had a longer duration of surgery but less intraoperative bleeding volume and more postoperative drainage fluid volume and retrieved lymph nodes compared to D2 radical surgery. LCME surgery is comparable to D2 surgery in terms of postoperative prognosis, but LCME surgery shows a positive trend in the overall survival curve. Show more

Keywords: Laparoscopic complete mesocolic excision, traditional D2 radical surgery, Colonic Neoplasms, propensity score matching

DOI: 10.3233/THC-241149

Citation: Technology and Health Care, vol. 33, no. 1, pp. 449-462, 2025

Authors: Hua, Defeng | Chen, Yan

Article Type: Research Article

Abstract: BACKGROUND: The present study investigated the association between cerebrovascular diseases and sepsis, including its occurrence, progression, and impact on mortality. However, there is currently a lack of predictive models for 28-day mortality in patients with cerebrovascular disease associated with sepsis. OBJECTIVE: The objective of this study is to examine the mortality rate within 28 days after discharge in this population, while concurrently developing a corresponding predictive model. METHODS: The data for this retrospective cohort study were obtained from the MIMIC-IV database. Patients with sepsis and cerebrovascular disease in the ICU were included. Laboratory …indicators, vital signs, and demographic data were collected within 24 hours of ICU admission. Mortality rates within 28 days after discharge were calculated based on patient death times. Logistic regression analysis was used to identify potential variables for a predictive model. A nomogram visualized the prediction model. The performance of the model was evaluated using ROC curves, Calibration plots, and DCA. RESULTS: The study enrolled a total of 2660 patients diagnosed with cerebrovascular disease complicated by sepsis, consisting of 1434 males (53.91%) with a median age of 70.97 (59.60, 80.73). Among this cohort of patients, a total of 751 fatalities occurred within 28 days following discharge. The multivariate regression analysis revealed that age, creatinine, arterial oxygen partial pressure (Pa O2), arterial carbon dioxide partial pressure (Pa CO2), respiratory rate, white blood cell (WBC) count, Body Mass Index (BMI), and race demonstrated potential predictive variables. The aforementioned model yielded an area under the ROC curve of 0.744, accompanied by a sensitivity of 66.2% and specificity of 71.2%. Furthermore, both calibration plots and DCA demonstrated robust performance in practical applications. CONCLUSION: The proposed prediction model allows clinicians to promptly assess the mortality risk in patients with cerebrovascular disease complicated by sepsis within 28 days after discharge, facilitating early intervention strategies. Consequently, clinicians can implement additional advantageous medical interventions for individuals with cerebrovascular disease and sepsis. Show more

Keywords: Predictive model, 28-day mortality, sepsis, cerebrovascular disease

DOI: 10.3233/THC-241150

Citation: Technology and Health Care, vol. 33, no. 1, pp. 463-472, 2025

Authors: Jundaeng, Jarupat | Chamchong, Rapeeporn | Nithikathkul, Choosak

Article Type: Research Article

Abstract: BACKGROUND: Artificial intelligence (AI) acts as the state-of-the-art in periodontitis diagnosis in dentistry. Current diagnostic challenges include errors due to a lack of experienced dentists, limited time for radiograph analysis, and mandatory reporting, impacting care quality, cost, and efficiency. OBJECTIVE: This review aims to evaluate the current and future trends in AI for diagnosing periodontitis. METHODS: A thorough literature review was conducted following PRISMA guidelines. We searched databases including PubMed, Scopus, Wiley Online Library, and ScienceDirect for studies published between January 2018 and December 2023. Keywords used in the search included “artificial intelligence,” …“panoramic radiograph,” “periodontitis,” “periodontal disease,” and “diagnosis.” RESULTS: The review included 12 studies from an initial 211 records. These studies used advanced models, particularly convolutional neural networks (CNNs), demonstrating accuracy rates for periodontal bone loss detection ranging from 0.76 to 0.98. Methodologies included deep learning hybrid methods, automated identification systems, and machine learning classifiers, enhancing diagnostic precision and efficiency. CONCLUSIONS: Integrating AI innovations in periodontitis diagnosis enhances diagnostic accuracy and efficiency, providing a robust alternative to conventional methods. These technologies offer quicker, less labor-intensive, and more precise alternatives to classical approaches. Future research should focus on improving AI model reliability and generalizability to ensure widespread clinical adoption. Show more

Keywords: Artificial intelligence, panoramic radiograph, periodontitis, periodontal disease, diagnosis

DOI: 10.3233/THC-241169

Citation: Technology and Health Care, vol. 33, no. 1, pp. 473-484, 2025

Authors: Shao, Liang | Gao, Ying | Zhang, Dan | Yang, Mengdan | Jiang, Mimi | Li, Hongfeng | Yan, Yanting

Article Type: Research Article

Abstract: BACKGROUND: non-small cell lung cancer (NSCLC) accounts for more than 80% of all lung cancer populations. Stereotactic radiotherapy (SBRT) is mainly suitable for early NSCLC patients who are not suitable for surgery or refuse surgery. OBJECTIVE: To analyze the effects of stereotactic radiotherapy (SBRT) plus immunotherapy for non-small cell lung cancer (NSCLC) patients on their immune status and survival quality. METHODS: NSCLC patients admitted to our hospital from 2019–2022 were divided into 61 cases in control group (SBRT) and 60 cases in observation group (SBRT plus immunotherapy) by the randomized numerical table method …to compare the efficacy, the level of tumor markers in the serum, the level and activity of the immune cells in the peripheral blood and the Kahlil’s functional status (KPS) scores. RESULTS: The observation group had a higher efficacy rate than that of the control group (P < 0.05). There was no statistical difference between the two groups in serum tumor marker content, immune cell level and activity in peripheral blood and KPS score before treatment (P > 0.05). After treatment, serum tumor markers were lower than those in control group, and immune cell level, NK cell-related activity and KPS score were higher than those in control group (P < 0.05). CONCLUSION: SBRT plus immunotherapy can reduce the level of various tumor markers, improve the immune status and quality of survival for NSCLC patients. Show more

Keywords: Stereotactic radiotherapy, immunotherapy, non-small cell lung cancer, immune status, survival quality

DOI: 10.3233/THC-241177

Citation: Technology and Health Care, vol. 33, no. 1, pp. 485-494, 2025

Authors: Han, Yongmei | Liu, Xinyu | Meng, Xinwan

Article Type: Research Article

Abstract: BACKGROUND: This study aimed to investigate the effects of the Kidney-Tonifying and Blood-Activating Formula on combating the downregulation of integrin β 3 and integrin β 5 in mouse placental tissue induced by phospholipid antibodies. OBJECTIVE: This study aimed to investigate the effects of kidney-tonifying and blood-activating formulations on combating the downregulation of integrin β 3 and integrin β 5 in mouse placental tissue induced by phospholipid antibodies. METHODS: Mice in the phospholipid antibody group and phospholipid antibody + kidney-tonifying and blood-activating formula …group underwent repeated implantation experiments, and the expression of integrin β 3 and integrin β 5 in placental tissue was observed. The effects of the kidney-tonifying and blood-activating formula on mouse placental tissue were evaluated through biochemical index tests, histopathological observations, and immunohistochemical staining. RESULTS: After intervention with the kidney-tonifying and blood-activating formula, the expression of integrin β 3 and integrin β 5 in placental tissue was notably upregulated, and placental structure was restored. Kidney-Tonifying and Blood-Activating Formula significantly improved abnormal biochemical indices and exerted a remarkable protective effect on placental tissue pathology. CONCLUSION: The kidney-tonifying and blood-activating formula effectively counteracted the downregulation of integrin β 3 and integrin β 5 in mouse placental tissue induced by phospholipid antibodies and alleviated placental tissue pathology through structural improvement and protection. Therefore, kidney-tonifying and blood-activating formulas may serve as potential therapeutic agents for recurrent implantation failure, offering new insights and approaches for clinical treatment. Show more

Keywords: Kidney-tonifying and blood-activating formula, phospholipid antibodies, recurrent implantation failure, placental tissue, integrin β3, integrin β5

DOI: 10.3233/THC-241179

Citation: Technology and Health Care, vol. 33, no. 1, pp. 495-507, 2025

Authors: Ji, Yanping | Ding, Qingbin | Bu, Tianyi | Wang, Kun | Zhang, Jing | Wang, Xue

Article Type: Research Article

Abstract: BACKGROUND: Cirrhosis is one of the most common clinical diseases in the world. The patient with cirrhosis has many complications, such as spontaneous peritonitis, hepatic encephalopathy. Malnutrition is one of the most common complications in patients with cirrhosis, and it is also a key factor affecting the prognosis of patients. However, the research on cirrhosis malnutrition is relatively scarce. OBJECTIVE: This study aimed to explore the new factors of the liver cirrhosis with the nutritional status. METHODS: A total of 370 patients with liver cirrhosis were admitted to the Fourth Affiliated Hospital of …Harbin Medical University from January 2019 to January 2023 were selected. Patients were assigned to the malnourished group and normal group. The weight, height, mid arm circumference (MAC) and triceps skinfold thickness (TSF) of the two groups were measured. Body mass index (BMI) and mid-arm muscle circumference (MAMC) were calculated. Furthermore, the Health Literacy Management Scale (HeLMS), biochemical indexes, incidence of complications, disease grade were also counted and the Royal Free Hospital-Nutrition Prioritization Tool were used to evaluate the nutritional status of the patient. RESULTS: A total of 370 patients was included in this study, including 177 malnutrition patients, accounting for 45.29%, and the scores and total scores of the patients in the malnutrition group were lower than those in the normal group. The measured values of Na+ , Alb, PA, T-Bil, TC, TLC, Hb, and RBC were all lower than those of the normal group, and the difference was statistically significant. The overall complication rate of the patients in the malnourished group was 89.61%, and that of the normal group was 39.78%. Multifactorial logistic regression analysis was performed with nutritional level as the dependent variable (normal = 0, malnutrition = 1), and the above indicators of variability (Na+ , Alb, PA, T-Bil, TC, PT, Hb, RBC, HG, and HeLMS scores) as the independent variables. The results showed that Na+ , Hb were the influencing factors of nutritional level (P < 0.05). CONCLUSION: The incidence of malnutrition in cirrhotic patients included in this study was at a moderate level, and the nutritional literacy of these patients was low. In addition, the level of serum sodium, Hemoglobin can affect the nutritional level of patients with cirrhosis. Show more

Keywords: Cirrhosis, malnutrition, correlation factor

DOI: 10.3233/THC-241180

Citation: Technology and Health Care, vol. 33, no. 1, pp. 509-518, 2025