Multi-task learning (MTL) has demonstrated significant advantages in the field of physiological signal measurement. This approach enhances the model's generalization ability by sharing parameters and features between similar tasks, even in data-scarce environments. However, traditional multi-task physiological signal measurement methods face challenges such as feature conflicts between tasks, task imbalance, and excessive model complexity, which limit their application in complex environments. To address these issues, this paper proposes an enhanced multi-scale spatiotemporal network (EMSTN) based on Eulerian video magnification (EVM), super-resolution reconstruction and convolutional multilayer perceptron. First, EVM is introduced in the input stage of the network to amplify subtle color and motion changes in the video, significantly improving the model's ability to capture pulse and respiratory signals. Additionally, a super-resolution reconstruction module is integrated into the network to enhance the image resolution, thereby improving detail capture and increasing the accuracy of facial action unit (AU) tasks. Then, convolutional multilayer perceptron is employed to replace traditional 2D convolutions, improving feature extraction efficiency and flexibility, which significantly boosts the performance of heart rate and respiratory rate measurements. Finally, comprehensive experiments on the Binghamton-Pittsburgh 4D Spontaneous Facial Expression Database (BP4D+) fully validate the effectiveness and superiority of the proposed method in multi-task physiological signal measurement.
Objective To investigate the use of intraoperative transit time flow measurement （TTFM） to accuratelyevaluate graft patency during sequential coronary artery bypass grafting （CABG）. Methods Clinical data of 131 patientsundergoing sequential off-pump coronary artery bypass grafting （OPCAB） with the great saphenous vein （with or without internal mammary artery） as graft vessels in Beijing Anzhen Hospital from April 2012 to January 2013 were retrospectivelyanalyzed. There were 92 male and 39 female patients with their age of 61.35±8.24 years. During the operation，2 methods were used to measure mean blood flow volume，pulsatility index （PI） and diastolic filling （DF） of the graft vessels. For thenon-blocking method，blood flow in graft vessels was maintained，and TTFM was applied 2 cm proximal to the anastomoticsite in graft vessels to record above parameters. For the blocking method，blood flow in graft vessels was temporally blockedby clipping distal graft vessels with an atraumatic bulldog clamp，and TTFM was applied 2 cm proximal to the anastomotic site in graft vessels to record above parameters. Results Blood flow volumes of the diagonal branch （27.43±15.22 ml/minvs. 59.28±30.13 ml/min），obtuse marginal branch （26.14±19.74 ml/min vs. 47.19±24.27 ml/min） and posterior left ventr-icular branch （19.16±8.92 ml/min vs. 38.83±20.11 ml/min） measured by the blocking method were significantly smallerthan those measured by the non-blocking method （P＜0.05） . PI values of the diagonal branch （2.93±1.30 vs. 2.31±0.91），obtuse marginal branch （2.62±1.17 vs. 2.01±0.87） and posterior left ventricular branch （2.33±0.92 vs. 1.80±0.73） meas-ured by the blocking method were significantly higher than those measured by the non-blocking method （P＜0.05）. There was no statistical difference in DF values measured by 2 methods at respective anastomotic sites，and all DF values were higher than 50%. Non-blocking method identified 1 anastomotic site and blocking method identified 3 anastomotic sites that were not patent，and these anastomotic sites became patent after graft reconstruction. Conclusion Blood flow velocity of sequential grafts is higher than that of single grafts，indicating that sequential grafts have the advantages of high blood flow volume and low risk of thrombosis. Blocking method can more accurately evaluate graft patency at the anastomotic sites and sensitively identify graft vessels that are not patent，which is helpful for anastomotic site reconstruction in time and enhancement of successful rate of CABG.
Rapid development of recently emerging precision medicine techniques represented by gene therapy has brought hope for the treatment of rare blinding eye diseases such as inherited retinal diseases (IRDs) for which there was no effective treatment previously. Although the globally growth of clinical trials for IRDs has increased rapidly over the past decade, due to the highly genetic and clinical phenotypic heterogeneity, as well as limited data on epidemiology and natural history of the disease, along with severe loss of vision function of majority patients for which the established measurements may not be appropriate, such studies lack standard outcome measurements and endpoints to asses clinical meaningful effectiveness, posing great challenges in terms of study design and evaluation of treatment efficacy, as well as clinical practice application. At present, there is no systematic nor standardized guidance on safety measures, clinical outcomes and endpoints of visual function for clinical trial design in IRDs. Therefore, in order to standardize the validated evaluation of IRDs clinical efficacy outcome measurements and endpoints, the Fundus Disease Group of Chinese Medical Association Ophthalmology Branch and Fundus Disease Committee of Chinese Medical Doctor Association Ophthalmology Branch organized domestic experts to put forward consensus and recommendations on standardizing outcome measurements and endpoints for clinical study design in IRDs, aiming to advance the study design of IRDs natural history research and clinical trials and to effectively evaluate disease progression and intervention efficacy. Along with the development of medical science and clinical trials, relevant content will be improved and updated accordingly.
In order to quantitatively evaluate the performance of dry electrode for fabric surface bioelectricity, a set of active measuring devices that can simulate electrocardiosignal has been developed on the basis of passive system by our group. Five Ag/AgCl fabric dry electrodes were selected to test and evaluate the devices. The results show that the deviation ratios of peak time interval of the five electrodes are all less than 1%. The maximum voltage amplitude decay rate is 7.2%, and the noise amplitudes are lower than 0.004 mV. The variable coefficient of peak time offset is less than 8%. The variable coefficient of voltage amplitude is less than 2%. The variable coefficient of noise amplitude is less than 10%. Research shows the devices has good repeatability and stability in measuring the simulated electrocardiosignal. The active measuring devices proposed in this paper can provide a new method for performance evaluation and standard formulation of surface bioelectricity dry electrode.
Objective To evaluate precise assessment methods for predicting the optimal acetabular cup size in total hip arthroplasty (THA). Methods A clinical data of 73 patients (80 hips) who underwent primary THA between December 2022 and July 2024 and met the inclusion criteria was analyzed. There were 39 males and 34 females with an average age of 66.3 years (range, 56-78 years). Among them, 66 cases were unilateral THA and 7 were bilateral THAs. There were 29 patients (34 hips) of osteoarthritis, 35 patients (35 hips) of femoral neck fractures, and 9 patients (11 hips) of osteonecrosis of the femoral head. Based on anteroposterior pelvic X-ray films, three methods were employed to predict acetabular cup size, including preoperative template planning, radiographic femoral head diameter (FHD) measurement, and intraoperative FHD measurement. The predicted acetabular cup sizes from these methods were compared with the actual implanted sizes. Results The predicted acetabular cup sizes using the preoperative template planning, radiographic FHD measurement, and intraoperative FHD measurement were (51.25±2.81), (49.72±3.11), and (49.90±2.74) mm, respectively, compared to the actual implanted cup size of (50.57±2.74) mm, with no significant difference (P>0.05). Regarding agreement with the actual implanted cup size, the preoperative template planning achieved exact matches in 35 hips (43.75%), one-size deviation in 41 hips (51.25%), and two-size deviations in 4 hips (5%); the radiographic FHD measurement achieved exact matches in 12 hips (15%), one-size deviation in 57 hips (71.25%), and two-size deviations in 11 hips (13.75%); and the intraoperative FHD measurement achieved exact matches in 26 hips (32.5%), one-size deviation in 52 hips (65%), and two-size deviations in 2 hips (2.5%). There were significant differences in agreement distributions between the three methods and the actual implanted cup sizes (H=18.579, P<0.001). Conclusion The intraoperative FHD measurement, as a simple, cost-effective, and accurate method, effectively guides acetabular cup selection, reduces the risk of prosthesis wear, enhances postoperative joint stability.
The risk of vertebral cortical shell fracture increases with aging. However, it remains unclear how aging contributes to cortex fracture at present. The aim of this study is to make understanding of the mechanism of how the spinal aging influences the cortical shell strain. Two finite element (FE) models of spinal segments (mildly and fully aged) were created, and then were compared to the FE models of the healthy spinal segment. The FE models of the aged spinal segments were generated by modifying both the geometry of the intervertebral disc (IVD) and the material properties of the spinal components. To find out under which case the cortical shell strain was influenced more, we created two types of FE model comparison methods: one with changes only in the spinal material properties and the other with changes only in the IVD geometry. The results showed that the cortical shell strains increased with aging and that compared to the changes of IVD geometry, the changes of spinal material property have a higher influence on the cortical shell strains. This study may suggest that for the prevention and treatment of vertebral cortex fracture, the augmentation of the vertebral body is a more effective treatment.
Hypertension is the primary disease that endangers human health. A convenient and accurate blood pressure measurement method can help to prevent the hypertension. This paper proposed a continuous blood pressure measurement method based on facial video signal. Firstly, color distortion filtering and independent component analysis were used to extract the video pulse wave of the region of interest in the facial video signal, and the multi-dimensional feature extraction of the pulse wave was preformed based on the time-frequency domain and physiological principles; Secondly, an integrated feature selection method was designed to extract the universal optimal feature subset; After that, we compared the single person blood pressure measurement models established by Elman neural network based on particle swarm optimization, support vector machine (SVM) and deep belief network; Finally, we used SVM algorithm to build a general blood pressure prediction model, which was compared and evaluated with the real blood pressure value. The experimental results showed that the blood pressure measurement results based on facial video were in good agreement with the standard blood pressure values. Comparing the estimated blood pressure from the video with standard blood pressure value, the mean absolute error (MAE) of systolic blood pressure was 4.9 mm Hg with a standard deviation (STD) of 5.9 mm Hg, and the MAE of diastolic blood pressure was 4.6 mm Hg with a STD of 5.0 mm Hg, which met the AAMI standards. The non-contact blood pressure measurement method based on video stream proposed in this paper can be used for blood pressure measurement.
To realize the accurate positioning and quantitative volume measurement of tumor in head and neck tumor CT images, we proposed a level set method based on augmented gradient. With the introduction of gradient information in the edge indicator function, our proposed level set model is adaptive to different intensity variation, and achieves accurate tumor segmentation. The segmentation result has been used to calculate tumor volume. In large volume tumor segmentation, the proposed level set method can reduce manual intervention and enhance the segmentation accuracy. Tumor volume calculation results are close to the gold standard. From the experiment results, the augmented gradient based level set method has achieved accurate head and neck tumor segmentation. It can provide useful information to computer aided diagnosis.
In this paper, a new probe is proposed for the in vivo dielectric measurement of anisotropic tissue in radio frequency band, which could accomplish the dielectric measurement in perpendicular directions by one operation. The simulative studies are performed in the frequency range from 1–1 000 MHz in order to investigate the influence of probe dimension on the energy coupling and sensitivity of measurement. The suitable probe is designed and validated for the actual measurement in this frequency band. According to the simulation results, the energy coupling of the probe could be kept below –12 dB in the frequency range from 200–400 MHz with high sensitivity of measurement for the dielectric properties of anisotropic tissue. That indicates the new type of probe has the potential to achieve the dielectric measurement of anisotropic tissue in radio frequency band and could avoid the measurement error by multi-operations in the conventional method. This new type of probe could provide a new method for the in vivo dielectric measurement of anisotropic tissue in radio frequency band.
Stage ⅠA lung adenocarcinoma presented as ground glass dominant on thin-section high-resolution CT scan is a special subtype of lung cancer. The characteristics of this subtype are quite different from the other patients, which presented as lower malignancy and better prognosis. Clinical, pathological and imaging studies have revealed that the proportion of the solid component in part-solid ground glass nodule is closely related with the pathological type and the prognosis of lung cancer. The methods for the assessment of the solid components in the ground glass nodule can be classified into three types, including subjective assessment, two dimensional measurement and three dimensional measurement. This review summarized the advantages and the limitations of these three methods. We also reviewed the clinical application of these techniques.