Ebook: Computer Methods in Medicine and Health Care
The last decade has seen great progress in modern healthcare; progress which could not have been achieved without the developments in computer science and technology. Image processing, disease modeling, biosensors and bioprinting are just some of the innovations which have contributed to improving the accuracy and efficiency of diagnosis and the more effective treatment of patients.
This book presents the accepted papers from CMMHC2022, the Workshop on Computer Methods in Medicine & Health Care, hosted from Hainan, China and held as an online event from 22 to 25 September 2022. CMMHC is aimed at fostering high-quality research by bringing together scholars, doctors, engineers, and radiologists to discuss emerging ideas, approaches, theories, frameworks, and practices in preventive healthcare technology. The organizers received 40 submissions for the 2022 workshop. These were subjected to a thorough peer-review process, with each paper being reviewed by at least 2 members of the Technical Committee. From the original submissions, 15 were selected for presentation and publication, resulting in a final acceptance rate of less than 40%. The main focus of the papers is hospital informatisation, medical imaging and health management, which continue to be major research hotspots.
The book offers an overview of recent research and developments in the field of computer methods in medicine and healthcare, and will provide a useful reference to the direction of future developments for those researchers and practitioners facing the challenges and demands of the era of big data.
This collection includes accepted papers from the 2022 Workshop on Computer Methods in Medicine & Health Care (CMMHC 2022), held from Sept 22–25th, 2022. CMMHC 2022 is one of the TDI series of conferences and was a continuation of CMMHC 2021. Its aim was to bring together scholars, doctors, engineers, and radiologists to discuss the recent status of research with the aim of directing future developments.
Forty researchers from universities, laboratories and hospitals submitted papers to CMMHC 2022, the majority focused on hospital informatization, medical imaging and health management, revealing the latest research hotspots, which receive extensive attention.
Special thanks are due to the 18 reviewers of the Technical Committee who participated in the peer-review of the 40 submissions, with at least two reviewers assigned to each paper. Fifteen papers were selected after peer-review, resulting in a final acceptance rate of less than 40%.
We would also like to thank all those authors who submitted a paper to CMMHC 2022 and followed the review suggestions. In accordance with the local regulations, the continuing impact of the Coronavirus prevented us from meeting face-to-face at Hainan University in 2022, but researchers from various countries were welcomed on ZOOM.
The editors and authors hope that this collection will provide a useful reference for those researchers and practitioners facing the challenges and meeting the demands of the era of big data.
Finding new ways to prevent and reduce the incidence of dementia is a serious world problem. This study aimed to perform imaging comparisons between pre- and post-meridian sinew therapy using arterial spin labeling (ASL) and resting-state functional MRI (rs-fMRI). Meanwhile, the results were studied to provide imaging evidence to support the effect of this meridian sinew therapy to slow down the brain aging and to reveal the related neurological mechanisms. Eighteen sub-healthy volunteers were selected as subjects. Three treatment strategies were adopted, acupuncture (group A), myofascial release (group B), and the integrated acupuncture and myofascial release (group C). The subjects were assigned to receive the three treatment modalities sequentially. 3T MRI examinations were provided before and after each treatment, including routine brain MRI plain scan, ASL and rs-fMRI scan. Compared with the results before and after treatment, the number of brain regions with increased cerebral blood flow (CBF) values in the group A, group B, and group C were respectively 1, 15 and 10 brain regions, all including the right cingulate gyrus. And rs-fMRI showed that multiple brain regions was activated, mainly temporal lobe and frontal lobe. The independent component analysis showed that the right intraorbital superior frontal gyrus and the occipital region was activated. Meridian sinew therapy can increase CBF and enhance neuronal activity in brain regions significantly associated with cognitive and memory functions, which may be the main targets where it actions on to achieve “Xingshen Yizhi (waking up the spirit and reinforcing thinking activity)” effect. The combination of ASL and rs-fMRI may be an effective imaging modality for future quantitative monitoring of the preventive and therapeutic effects of the meridian sinew therapy.
The aim of this study is exploring the value of diffusional kurtosis imaging (DKI) in the diagnosis of acute carbon monoxide poisoning encephalopathy in vivo rats. Forty healthy male Sprague-Dawley rats, were divided into a control group and CO poisoning model group. DKI during 7.0 T MR were performed in the globus pallidus, hippocampus and parietal cortex in the rats. During acute CO poisoning, compared with controls, Mean kurtosis values (MK) and Mean diffusivity (MD) values were significantly decreased both in the hippocampus and parietal cortex, however, in the globus pallidus were significantly increased.The pathological findings showed emerged granular cytoplasmic changes, thickening of chromatin of the neurons, and local lymphocytic infiltration. However, the condition in the hippocampus and parietal cortex was significantly milder than in the globus pallidus region in the first day or 3 days later. Both Immunohistochemical analysis of the heme oxygenase-1 of rats and MK values increased only in globus pallidus with a statistical difference between the CO poisoning group and the normal group both in the day 1 and day 3. DKI can possess sufficient sensitivity for tracking pathophysiological changes associated with carbon monoxide intoxication. The higher MK values in acute stage of carbon monoxide intoxication might indicate poor prognosis in the evolution of the condition.
The therapeutic elastic gloves are of great help in the treatment of hand burns and scalds, making superficial and mild burns heal quickly without residual scars or damage to a patient’s functional ability. However, the hand data used to create the elastic gloves for burns and scalds are usually measured by workers. This manual measurement method has a high cost, large error, and unreliable results. In this paper, we propose an image-based parameter measurement method and establish a portable measuring system for finger and palm parameters; these parameters are then applied to create therapeutic gloves for burn and scald treatments. The proposed method can provide an accurate and rapid measurement of the finger and palm parameters. The experimental results for normal hand parameters and injured hand parameters show the effectiveness of the proposed method.
To solve the problem that the current deep learning method is difficult to deal with the recognition of nested entities in Chinese medical text, a deep learning model based on word-word relationship is introduced, and the relationship between words is built by multi-granularity 2D graphs to improve the recognition of nested entities. First, we use BERT (Bidirectional Encoder Representation from Transformers) for pre-training, then we use BiLSTM (directional Long Short-Term Memory) to extract the context information. Then, we merge the token representation information, the word distance information and the word regional information, through use a multi-granularity hole convolution to obtain the role information of different words. Finally, we use decoding layer to predict entity relationships and decode the result. This model is tested on the CMeEE Chinese medical dataset. Compared with the popularity models such as BiLSTM-CRF (Conditional Random Field) and BERT-BiLSTM-CRF, the F1 value is improved by 2.52%. Experimental results show that for Chinese medical named entity recognition with nested entities, this model can better recognize the medical entities in Chinese medical text.
To evaluate the experimental effect of resistance exercise on health management of exercise therapy in elderly female patients with osteoporosis. 79 patients with osteoporosis were divided into the conventional treatment group, aerobic exercise group, and resistance training group for the experimental study. Results: After 4 months of treatment, only the difference of L2 to L4 vertebral bone density in resistance training was statistically significant, after 8 months of treatment, there were statistically significant differences in BMD of foot root bone, radius bone, and L2 ∼ L4 vertebrae in the three groups, and the added BMD of resistance training was greater than that of the control group and the aerobic exercise group. Conclusion: (1) All three groups have a therapeutic effect on osteoporosis in middle-aged and elderly women. the therapeutic effect of resistance training is significantly better than that of the other two groups. The therapeutic effect is shown as follows: resistance training group > aerobic exercise group > routine treatment group. (2) The BMD increase at 12 months was greater than that at 8 months and 4 months, indicating that the longer the treatment, the more obvious the effect.
The contactless service has become an important measure to avoid the spread of COVID-19 since its outbreak in 2020. Combining with the current epidemic situation and the epidemic prevention problems exposed by the present situation of outpatient service in hospitals and the analysis of their causes, this study applied the thinking and method of product design, and put forward an intelligent medical guide service robot design. By analyzing the user’s actual outpatient demand, the functional modules of the robot are planned, and the corresponding hospital scene is established by using the open source simulation software Webots, which simulates the navigation function of the robot. The purpose of this study is to reduce the labor burden of medical staff, alleviate the contradiction of medical resources shortage, improve the quality of hospital outpatient service and build a new intelligent medical service platform. Meanwhile, it also provides relevant reference for the design of intelligent medical products.
Due to the increasing pressure of social competition, the fierce conflicts of various values during the social transformation, the complex interpersonal relationships, and the continuous and normalized development of the epidemic, etc., the emergence of mental health problems of college students has been exacerbated. And the various psychological symptoms produced in the process of life continue to increase. The mental health problems of college students have a deepening trend, becoming an important social problem. In response to this situation, we must come up with corresponding solutions to effectively guide and control the various adverse effects of college students due to mental health problems. To this end, a psychological consultation system is designed, and colleges and universities can use this system to provide psychological consultation for students. In the system design, the PSO algorithm of the intelligent optimization algorithm is introduced to optimize the system performance, and through the improvement of the performance, it can better provide consulting services for students. Comparing the performance before and after optimization of the algorithm proves the effectiveness of the PSO algorithm in improving the performance. The results of the system satisfaction survey also verify that the system in this paper meets the needs of students for psychological education counseling.
In this article, we apply the stability theory of differential equations, based on the improved infectious disease transmission model SEIS, to describe the change in the number of infections when the lurker is a non-staff. In the process of the spread of infectious diseases, we establish the relationship between various groups, and establish the equation data solving algorithm. On this basis, a complex network model is established to describe the influence of the movement of various groups of people in the system on the number of infections when the lurker is a staff member. At the same time, the cellular automata simulation in accordance with the complex network models is carried out through the collected data. Finally, using the probabilistic model of the spread of infectious diseases, the impact of the protective effect on the spread of infectious diseases is analyzed when staff in public places take appropriate protective measures. Through the establishment of the probabilistic model and the curve fitted by the python program, we conclude that at the beginning of the spread of infectious diseases, the fastest and best protective measures can not only slow down the speed of the spread of infectious diseases, but also effectively reduce the infection in the later stages of transmission the proportion of the people.
Classification problems are important in medical diagnosis. In this workshop, we present and summarize our recent insights into AI uncertainty through classification problems from both theoretical and empirical perspectives. First a concept and a theory are proposed aiming at zero-error AI system. We show in fact they can be derived from Shannon communication theory based on entropy concept. Classical Rademacher complexity and Shannon entropy is shown to be closely related by quantity by definitions. Based on this observation, we are able to derive a 1/2 criteria in terms of Shannon entropy that guarantee an AI zero-error accuracy in classifications problems. Last but not the least, we show both a relaxing condition and a stricter condition in real applications that can guarantee zero-error accuracy in AI classification problems. We provide examples to show in applications how to apply our derived 1/2 criteria into AI applications by “coding” methods.
After experiencing problems such as shutdown and limited offline activities caused by the COVID-19 pandemic, various industries have taken the initiative to take self-help measures such as Internetization, which makes the industry economy change significantly in the post-pandemic era and makes it necessary to re-evaluate the industry economic pattern. However, online public opinion is diverse and platform content is complex. Therefore, it is essential to observe economic activities from the network information platform and complete the screening and purification of information. In order to improve the degree of quantification, and deepen the understanding of data, this paper solves the above problems through parallel coordinate visualization. At the same time, the lines in the parallel axis are used to indicate the amount and specific trend of industry information, which not only reflects the role of parallel coordinate visualization methods in industry dynamic analysis and real-time display but also makes industry development forecasts more feasible.
This paper analyzes the current situation of hospital informatization, and puts forward a new solution to realize the integration of hospital information system to make up the defects of the traditional integration scheme in the current hospital informatization construction. Furthermore, it introduces the software architecture and development planning, as well as expounds the operation process of Internet hospital, including website diagnosis and treatment and Medical Union service system. Each independent medical information subsystem is closely linked through the hospital information integration platform. On the base of the open architecture, the integration platform has strong scalability and supports the connection of customized adapters and the application of complex heterogeneous systems, so as to realize the data sharing and information collaboration among various hospital systems.
Computer science and technology have gone deep into many fields, including medical care and public health. The network teaching platform, online courses and formative evaluation developed based on computer science and technology have been widely used in the teaching process of public health, and have played a key role in training qualified public health talents. Case-based learning is a teaching method that takes students as the center and cases as the basis, in which students are guided to find, analyze and solve problems by presenting case situations, and closely combining theory with practice, so that they can understand theories, form their own views and improve their abilities. In this study, the necessity and importance of case-based learning, the basic strategy for the construction of case database, and the design of typical cases of food toxicology were investigated in combination with the training objectives of public health major and the basic situation of food toxicology course in our university. At the same time, a case-based learning was applied in the course of food toxicology, and it was found that this teaching mode could effectively improve the students’ academic performance and teaching satisfaction, and should be of great help to students’ ability to obtain information, learn independently and evaluate information, worth popularizing.
In the face of public health emergencies, medical libraries provide various types of intelligent knowledge services for relevant users based on the available resources and conditions, and take active actions, which is of great significance for medical libraries to fulfill their original mission and responsibilities with practical actions, expand their business scope and innovate service forms. The intelligent service of medical library for public health emergencies is a typical complex system engineering, which requires the cooperation of staff with different professional knowledge and skills. According to a series of systematic processes, intelligent service uses various types of hardware and software resources to realize intelligent knowledge service and decision support. Based on the WSR system methodology in the field of systems engineering, this paper constructs a medical library’s intelligent service model for public health emergencies. The model is a three-dimensional space structure composed of the physical (Wu in Chinese) dimension (W), the common sense (Shi in Chinese) dimension (S), and the human (Ren in Chinese) dimension (R). The article analyzes and discusses the functions and implementation paths of each dimension in detail.
With the rapid improvement and wide application of network technology and the maturity of network tools, the network course platform has become a popular direction of teaching research in colleges and universities. The purpose of the psychological sand table online course is to enable students to achieve self-catharsis, release pressure and reflect self-growth through sand table and sand tools. The research purpose of this paper is to design the network platform of psychological sandbox course based on data mining. In the experiment, the sandplay course designed a 12-hour course corresponding to the mental health education course. Taking the students of Class 1 and Class 2 of the psychology education major of Wuyi University in Fujian Province as the research object, a questionnaire survey and an independent sample T test were carried out on the class. Therefore, the group psychological sandplay game course in colleges and universities is carried out, and the evaluation of the implementation effect of the psychological sandplay course network platform is studied in the experiment.
Lung cancer is a high incidence disease, which seriously affects people’s health. The pathological section of lung cancer can determine the type and differentiation of lung cancer cells, so as to provide an important basis for the selecting treatment options. In recent years, researchers focus on using convolutional neural network (CNN) algorithm to assist doctors and improve the recognition of cancerous regions in pathological images. In this paper, the CNN models were used to identify the cancerous region of lung cancer pathological images. The public data set was selected to train the AlexNet, GoogLeNet and ResNet34 models, and adjust the relevant parameters to improve the accuracy and specificity of recognition as much as possible. The experimental results showed that the accuracy and specificity of ResNet34 model were 98.9% and 99.0%, respectively, indicating that the model could effectively assist doctors to identify cancerous regions in lung cancer pathological images.