1.Research on multi-scale convolutional neural network hand muscle strength prediction model improved based on convolutional attention module.
Yihao DU ; Mengyu SUN ; Jingjin LI ; Xiaoran WANG ; Tianfu CAO
Journal of Biomedical Engineering 2025;42(1):90-95
In order to realize the quantitative assessment of muscle strength in hand function rehabilitation and then formulate scientific and effective rehabilitation training strategies, this paper constructs a multi-scale convolutional neural network (MSCNN) - convolutional block attention module (CBAM) - bidirectional long short-term memory network (BiLSTM) muscle strength prediction model to fully explore the spatial and temporal features of the data and simultaneously suppress useless features, and finally achieve the improvement of the accuracy of the muscle strength prediction model. To verify the effectiveness of the model proposed in this paper, the model in this paper is compared with traditional models such as support vector machine (SVM), random forest (RF), convolutional neural network (CNN), CNN - squeeze excitation network (SENet), MSCNN-CBAM and MSCNN-BiLSTM, and the effect of muscle strength prediction by each model is investigated when the hand force application changes from 40% of the maximum voluntary contraction force (MVC) to 60% of the MVC. The research results show that as the hand force application increases, the effect of the muscle strength prediction model becomes worse. Then the ablation experiment is used to analyze the influence degree of each module on the muscle strength prediction result, and it is found that the CBAM module plays a key role in the model. Therefore, by using the model in this article, the accuracy of muscle strength prediction can be effectively improved, and the characteristics and laws of hand muscle activities can be deeply understood, providing assistance for further exploring the mechanism of hand functions .
Neural Networks, Computer
;
Humans
;
Hand Strength/physiology*
;
Support Vector Machine
;
Muscle Strength/physiology*
;
Hand/physiology*
;
Convolutional Neural Networks
2.Clinical analysis of 54 cases of lung cancer treated by domestic carbon ion system
Xin PAN ; Yihe ZHANG ; Tong MA ; Xin WANG ; Yuling YANG ; Tianyan QIN ; Caixia LYU ; Pengqing LI ; Yancheng YE ; Yanshan ZHANG
Chinese Journal of Radiation Oncology 2024;33(4):319-325
Objective:To evaluate clinical prognosis and prognostic factors of patients with early stage (Ⅰ stage) and locally advanced (Ⅱ/Ⅲ stage) lung cancer treated with carbon ion radiotherapy (CIRT).Methods:Clinical data, treatment, adverse reactions, survival and so on of 54 lung cancer patients who received CIRT and follow-up in the Heavy Ion Center of Wuwei Cancer Hospital of Gansu Province from March 2020 to September 2022 were retrospectively analyzed. The survival curve was plotted using Kaplan-Meier method. Difference tests were performed using log-rank test. Logistic regression analysis was used to identify prognostic factors.Results:According to inclusion and exclusion criteria, 54 patients were enrolled in the study, including 10 patients with early stage lung cancer and 44 patients with locally advanced lung cancer. The median follow-up time for 10 patients with early stage lung cancer was 11.0 (6.75, 17.25) months, and the median dose of irradiation was 60 Gy [relative biological effect (RBE)]. Upon the last follow-up, 3 patients had complete response (CR) and 3 patients had partial response (PR). Four patients had stable disease (SD) and no progressive disease (PD). The 1-year and 2-year local control rates (LCR), progression-free survival (PFS) rates and overall survival (OS) rates were 100%. During treatment and follow-up, 2 patients developed grade 1 radiation pneumonia, 1 case of grade 2 radiation pneumonia, 1 case of chest wall injury (chest wall pain), and there were no adverse reactions greater than grade 2. The median follow-up time of 44 patients with locally advanced stage was 12.5 (4.25, 21.75) months, and the median irradiation dose was 72 Gy (RBE). Thirty-two (73%) patients received concurrent chemotherapy during treatment, 20 (45%) patients received sequential chemotherapy after treatment, 14 (32%) patients received immune maintenance therapy and 3 (7%) patients obtained PD and received targeted drugs. Upon the last follow-up, 3 (7%) patients had CR, 17 (39%) patients had PR, 19 (43%) patients obtained SD, and 5 (11%) patients had PD. The 1-year and 2-year LCR were 96.0% and 87.3%, 90.9% and 84.1% for the 1-year and 2-year PFS rates, and 93.2% and 86.4% for the 1-year and 2-year OS rates, respectively. The median OS and PFS of patients were not reached. Multivariate logistic regression analysis showed that maintenance therapy after radiotherapy ( P=0.027) and clinical target volume (CTV) irradiation volume ( P=0.028) were the factors affecting PFS. Simultaneous chemoradiotherapy ( P=0.042) and maintenance therapy after radiotherapy ( P=0.020) were the factors affecting OS. And gross tumor volume (GTV) ≥215 ml ( P=0.068) might be an independent risk factor for grade 2 and above radiation pneumonia. Conclusions:The domestic carbon ion system has definite clinical effect and controllable toxic and side effects in the treatment of early stage and locally advanced lung cancer. The combination of synchronous chemotherapy and further maintenance treatment can significantly improve clinical prognosis of patients without significantly increasing the risk of toxic and side effects.
3.Influencing factors of anxiety symptoms in firstborn preschool children
Aimei YE ; Feng CHEN ; Yuzhong YE ; Changcan HUANG ; Junmin LI ; Yanshan WANG ; Dongxi LU ; Mujin GUO ; Weige WU ; Xiaoling LIN ; Dali LU
Sichuan Mental Health 2024;37(6):537-542
BackgroundSibling relationships play a critical role in shaping anxiety symptoms in firstborn children. Anxiety symptoms often originate in early childhood and can persist into adolescence and adulthood. However, there is insufficient research on anxiety symptoms in preschool children, especially firstborn preschool children. ObjectiveTo explore the influencing factors of anxiety symptoms among firstborn preschool children, so as to provide references for the intervention of anxiety symptom for children in families with multiple children. MethodsFrom October to December 2021, a total of 8 449 children from 234 kindergartens in Longhua District of Shenzhen were included using a cluster sampling method. Sibling Inventory of Behavior (SIB) and Spence Preschool Anxiety Scale (SPAS) were used to investigate. Logistic regression analysis was used to identify influencing factors of anxiety symptoms in firstborn preschool children. ResultsA total of 8 419 (99.64%) valid questionnaires were collected. Anxiety symptoms were detected in 344(4.09%) firstborn preschool children. Statistically significant differences were observed between anxiety group and non-anxiety group in terms of household registration, monthly family income, maternal age, maternal education level, paternal education level, family living conditions and whether they are left-behind children (χ2/t=9.906, 33.490, 5.136, 13.485, 9.690, 17.332, 21.975, P<0.05 or 0.01). Compared with non-anxiety group, children in the anxiety group scored higher on the SIB dimensions of rivalry, aggression and avoidance (t=165.322, 74.471, 286.419, P<0.01), and lower on companionship, empathy and teaching (t=59.133, 42.417, 39.112, P<0.01). Risk factors for anxiety symptoms in firstborn preschool children included left-behind children, as well as negative sibling relationships characterized by rivalry and avoidance (OR=1.195, 1.143, 1.260, P<0.05 or 0.01). ConclusionFirstborn preschool children who are left-behind are more susceptible to anxiety symptoms. Negative sibling relationships, characterized by competition and avoidance, may also contribute to the emergence of anxiety symptoms in firstborn preschool children.
4.The efficacy and safety of ibrutinib in the treatment of lymphoplasmacytic lymphoma/Waldenstr?m macroglobulinemia
Yanshan HUANG ; Wenjie XIONG ; Jingjing YUAN ; Ying YU ; Yuxi LI ; Yuting YAN ; Tingyu WANG ; Rui LYU ; Wei LIU ; Gang AN ; Yaozhong ZHAO ; Dehui ZOU ; Lugui QIU ; Shuhua YI
Chinese Journal of Hematology 2024;45(8):755-760
Objective:To explore the efficacy and safety of ibrutinib for the treatment of newly treated and relapsed refractory (R/R) lymphoplasmacytic lymphoma (LPL) /Waldenstr?m macroglobulinemia (WM) .Methods:Retrospectively collected clinical data of 98 cases of newly treated and R/R LPL/WM patients who received ibrutinib treatment at the Hematology & Blood Diseases Hospital of the Chinese Academy of Medical Sciences from March 2016 to June 2023, and analyzed their efficacy and safety.Results:A total of 98 LPL/WM patients were included, which consisted of 45 newly treated patients and 53 R/R patients. Of these, 74 were males (75.5%) and the cohort had a median age of 64 (42-87) years. Eighty-eight patients were eligible for efficacy evaluation with a median treatment time of 20.8 (2.1-55.0) months, a major remission rate (MRR) of 78.4%, and an overall response rate (ORR) of 85.2%. The MRR and ORR of the newly treated patients were 78.4% and 86.5%, respectively, whereas the MRR and ORR of the R/R patients were 78.4% and 84.3%, respectively. There were no statistically significant differences in MRR and ORR between the initial treatment and R/R patients (all P values >0.05) . The median follow-up period was 29.1 (2.9-50.3) months and the median overall survival time for newly treated and R/R patients was not reached. The median progression-free survival time was 23.5 (95% CI 10.5-36.5) months and 45.0 (95% CI 34.0-56.0) months, respectively, with no statistically significant differences (all P values >0.05) . There were 25 deceased patients and no deaths were related to ibrutinib treatment. The main adverse reactions of ibrutinib were thrombocytopenia (5.1%) , pneumonia (8.1%) , and hyperuricemia (21.4%) . The incidence of atrial fibrillation was 2.0%. Conclusion:Ibrutinib exhibits good efficacy and safety for newly treated and R/R LPL/WM patients.
5.Study on the regulatory effect of low intensity retinal ultrasound stimulation on the neural activity of visual cortex.
Qianqian WANG ; Yi YUAN ; Jiaqing YAN
Journal of Biomedical Engineering 2024;41(6):1161-1168
Low-intensity ultrasound stimulation of the retina has the ability to modulate neural activity in the primary visual cortex (V1), however, it is currently unclear how different intensities and durations of ultrasonic stimulation of the retina modulate neural activity in V1. In this paper, we recorded local field potential (LFP) signals in the V1 brain region of mice under different ultrasound intensities and different stimulation times. The amplitude of LFP corresponding to 1 s before ultrasound stimulation to 2 s after stimulation (-1-2 s) was analyzed, including the power and sample entropy of delta, theta, alpha beta, and low gamma frequency bands. The experimental results showed that, as the stimulation intensity increased, the peak value of the LFP in the visual cortex showed a linear upward trend; the power in the delta and theta frequency bands showed a linear upward trend, and the sample entropy showed a linear downward trend. With increases of stimulation duration, the peak value of the LFP in the visual cortex showed an upward trend, and the upward trend gradually weakened; the power in the delta frequency band showed an upward trend, the sample entropy showed a linear upward trend, and the sample entropy in the theta frequency band showed a downward trend. The results show that low-intensity ultrasonic stimulation of the retina has a significant modulatory effect on neural activity in the visual cortex. The study provides insights into the mechanisms by which ultrasonic stimulation regulates visual system function. Furthermore, it clarifies the patterns of parameter selection, facilitating the development of personalized multi-parameter modulation for the treatment of visual neural degeneration, retinal disorders and related research areas.
Animals
;
Visual Cortex/radiation effects*
;
Retina/radiation effects*
;
Mice
;
Ultrasonic Waves
;
Primary Visual Cortex/physiology*
6.Modulation of motor responses and neural activities with transcranial ultrasound stimulation based on closed-loop control.
Shuxun DONG ; Zhenyu XIE ; Xingran WANG ; Yi YUAN
Journal of Biomedical Engineering 2023;40(2):265-271
Closed-loop transcranial ultrasound stimulation technology is based on real-time feedback signals, and has the potential for precise regulation of neural activity. In this paper, firstly the local field potential (LFP) and electromyogram (EMG) signals of mice under different intensities of ultrasound stimulation were recorded, then the mathematical model of ultrasound intensity and mouse LFP peak/EMG mean was established offline based on the data, and the closed-loop control system of LFP peak and EMG mean based on PID neural network control algorithm was simulated and built to realize closed-loop control of LFP peak and EMG mean of mice. In addition, using the generalized minimum variance control algorithm, the closed-loop control of theta oscillation power was realized. There was no significant difference between the LFP peak, EMG mean and theta power under closed-loop ultrasound control and the given value, indicating a significant control effect on the LFP peak, EMG mean and theta power of mice. Transcranial ultrasound stimulation based on closed-loop control algorithms provides a direct tool for precise modulation of electrophysiological signals in mice.
Mice
;
Animals
;
Deep Brain Stimulation
;
Algorithms
;
Electromyography
7.Clinical report of two dose fractionation modes using carbon ion beam therapy in the lymph node drainage area for lung cancer
Xin PAN ; Yihe ZHANG ; Xiaojun LI ; Tong MA ; Xin WANG ; Yuling YANG ; Hongyu CHAI ; Tianyan QIN ; Caixia LYU ; Pengqing LI ; Yancheng YE ; Yanshan ZHANG
Chinese Journal of Radiation Oncology 2023;32(3):215-221
Objective:To compare the adverse reactions, efficacy and survival rate of carbon ion beam irradiation in the elective lymph node (ENI) drainage area of locally advanced non-small cell lung cancer (LA-NSCLC) with relative biological effect (RBE) dose of 48 Gy using 16 and 12 fractions.Methods:A total of 72 patients with pathologically confirmed LA-NSCLC admitted to Wuwei Heavy Ion Center of Gansu Wuwei Tumor Hospital from June 2020 to December 2021 were enrolled and simple randomly divided into groups A and B, with 36 patients in each group. Patients in groups A and B were treated with carbon ion beam irradiation to the lymph node drainage area with 48 Gy (RBE) using 16 and 12 fractions. The acute and chronic adverse reactions, efficacy and survival rate were observed. The survival curve was drawn by Kaplan-Meier method. Difference test was conducted by log-rank test.Results:The median follow-up time was 13.9 (8.8-15.7) months in group A and 14.6 (6.3-15.9) months in group B. Sixteen (44.4%) patients were effectively treated in group A and 9 (25%) patients in group B. Thirty-four (94.4%) cases achieved disease control in group A and 30 (83.3%) cases in group B. Statistical analysis showed that the overall survival rate in group B was similar to that in group A ( χ2=1.192, P=0.275). Comparison of planning parameters between two groups showed CTV volume, D mean, V 5 Gy(RBE), V 20 Gy(RBE) and V 30 Gy(RBE) of the affected lung, cardiac V 20 Gy(RBE), V 30 Gy(RBE) and D mean, esophageal V 30 Gy(RBE), V 50 Gy(RBE), D max and D mean, D max of the trachea and spinal cord had no significant difference (all P>0.05). No grade 3 or 4 adverse reactions occurred in the enrolled patients during treatment and follow-up. No statistical differences were observed in the acute radiation skin reaction ( χ2=5.134, P=0.077), radiation esophagitis ( χ2=1.984, P=0.371), and advanced radiation pneumonia ( χ2=6.185, P=0.103) between two groups. Conclusions:The two dose fractionation modes of carbon ion therapy system are equally safe in the mediastinal lymphatic drainage area of LA-NSCLC, and the adverse reactions are controllable. The long-term efficacy still needs further observation.
8.A study on the application of cross-frequency coupling characteristics of neural oscillation in the diagnosis of mild cognitive impairment.
Xin LI ; Kai WANG ; Jun JING ; Liyong YIN ; Ying ZHANG ; Ping XIE
Journal of Biomedical Engineering 2023;40(5):843-851
In order to fully explore the neural oscillatory coupling characteristics of patients with mild cognitive impairment (MCI), this paper analyzed and compared the strength of the coupling characteristics for 28 MCI patients and 21 normal subjects under six different-frequency combinations. The results showed that the difference in the global phase synchronization index of cross-frequency coupling under δ-θ rhythm combination was statistically significant in the MCI group compared with the normal control group ( P = 0.025, d = 0.398). To further validate this coupling feature, this paper proposed an optimized convolutional neural network model that incorporated a time-frequency data enhancement module and batch normalization layers to prevent overfitting while enhancing the robustness of the model. Based on this optimized model, with the phase locking value matrix of δ-θ rhythm combination as the single input feature, the diagnostic accuracy of MCI patients was (95.49 ± 4.15)%, sensitivity and specificity were (93.71 ± 7.21)% and (97.50 ± 5.34)%, respectively. The results showed that the characteristics of the phase locking value matrix under the combination of δ-θ rhythms can adequately reflect the cognitive status of MCI patients, which is helpful to assist the diagnosis of MCI.
Humans
;
Electroencephalography/methods*
;
Cognitive Dysfunction/diagnosis*
;
Neural Networks, Computer
;
Sensitivity and Specificity
9.The current applicating state of neural network-based electroencephalogram diagnosis of Alzheimer's disease.
Yi LIU ; Zhenyang LI ; Zhiwei WEI ; Yonghong XU ; Ping XIE ; Yulin WANG ; Qinshuang LIU ; Xin LI
Journal of Biomedical Engineering 2022;39(6):1233-1239
The electroencephalogram (EEG) signal is a general reflection of the neurophysiological activity of the brain, which has the advantages of being safe, efficient, real-time and dynamic. With the development and advancement of machine learning research, automatic diagnosis of Alzheimer's diseases based on deep learning is becoming a research hotspot. Started from feedforward neural networks, this paper compared and analysed the structural properties of neural network models such as recurrent neural networks, convolutional neural networks and deep belief networks and their performance in the diagnosis of Alzheimer's disease. It also discussed the possible challenges and research trends of this research in the future, expecting to provide a valuable reference for the clinical application of neural networks in the EEG diagnosis of Alzheimer's disease.
Humans
;
Alzheimer Disease/diagnosis*
;
Neural Networks, Computer
;
Machine Learning
;
Brain
;
Electroencephalography
10.Preliminary efficacy observation of carbon ion radiotherapy in the treatment of muscle-invasive bladder cancer in phase Ⅰ/Ⅱ clinical trials
Yanshan ZHANG ; Xiaojun LI ; Yihe ZHANG ; Xin PAN ; Xin WANG ; Yuling YANG ; Hongyu CHAI ; Tong MA
Chinese Journal of Radiological Medicine and Protection 2022;42(9):664-670
Objective:To evaluate the feasibility and safety of carbon ion radiotherapy (CIRT) in the treatment of muscle-invasive bladder cancer in phase Ⅰ/Ⅱ clinical trials.Methods:Clinical stage T 2-3 patients with muscle-invasive bladder cancer (without distant metastasis) were studied. A three-fraction treatment was applied, including the local irradiation with the dose from 12 Gy to 24 Gy and 11 fractions of whole-bladder irradiation with a dose of 44 Gy. The carbon ion irradiation dose is determined with relative biological effectiveness (RBE) of 3.0. The total dose for bladder tumor was 56-68 Gy in 14 fractions. The primary endpoints included tumor treatment-related side effects, dose-limiting toxicity (DLT) responses, and local control (LC) rate, and the secondary endpoints included progression-free survival (PFS). Results:Nine patients received CIRT of various doses in the clinical trials, with the dose gradually increasing to 68 Gy. The patients did not suffer from DLT response, acute adverse effects of radiation therapy of grade ≥3, and late radiation adverse reactions during follow-up. When the dose to the tumor reached 68 Gy, there were 2 cases of grade 2 acute urogenital tract reaction and 1 case of acute lower gastrointestinal tract symptom. For the group with a dose above 62 Gy, three cases of grade 1 late radiation bladder reaction were observed and their symptoms included urinary frequency and microscopic hematuria. At the end of treatment, hematuria disappeared, dysuria was relieved, and urine red blood cell value significantly decreased for all the patients. Three months and six months after treatment, the LC rates were 100% and 88.9%, respectively, and the objective response rates were both 88.9%. One patient developed local recurrence and was treated with salvage surgery six months after treatment.Conclusions:The preliminary efficacy observation of CIRT in the treatment of muscle-invasive bladder cancer showed significant short-term efficacy, obvious symptom relief, and good tolerability for patients, without DLT. Therefore, CIRT is safe and feasible.

Result Analysis
Print
Save
E-mail