ABSTRACT
Background: Deep Learning is an AI technology that trains computers to analyze data in an approach similar to the human brain. Deep learning algorithms can find complex patterns in images, text, audio, and other data types to provide accurate predictions and conclusions. Neuronal networks are another name for Deep Learning. These layers are the input, the hidden, and the output of a deep learning model. First, data is taken in by the input layer, and then it is processed by the output layer. Deep Learning has many advantages over traditional machine learning algorithms like a KA-nearest neighbor, support vector algorithms, and regression approaches. Deep learning models can read more complex data than traditional machine learning methods. Objectives: This research aims to find the ideal number of best-hidden layers for the neural network and different activation function variations. The article also thoroughly analyzes how various frameworks can be used to create a comparison or fast neural networks. The final goal of the article is to investigate all such innovative techniques that allow us to speed up the training of neural networks without losing accuracy. Methods: A sample data Set from 2001 was collected by www.Kaggle.com. We can reduce the total number of layers in the deep learning model. This will enable us to use our time. To perform the ReLU activation, we will make use of two layers that are completely connected. If the value being supplied is larger than zero, the ReLU activation will return 0, and else it will output the value being input directly. Results: We use multiple parameters to determine the most effective method to test how well our method works. In the next paragraph, we'll discuss how the calculation changes secret-shared Values. By adopting 19 train set features, we train our reliable model to predict healthcare cost's (numerical) target feature. We found that 0.89503 was the best choice because it gave us a good fit (R2) and let us set enough coefficients to 0. To develop our stable model with this Set of parameters, we require 26 iterations. We use an R2 of 0.89503, an MSE of 0.01094, an RMSE of 0.10458, a mean residual deviance of 0.01094, a mean absolute error of 0.07452, and a root mean squared log error of 0.07207. After training the model on the train set, we applied the same parameters to the test set and obtained an R2 of 0.90707, MSE of 0.01045, RMSE of 0.10224, mean residual deviation of 0.01045, MAE of 0.06954, and RMSE of 0.07051, validating our solution approach. The objective value of our secured model is higher than that of the scikit-learn model, although the former performs better on goodness-of-fit criteria. As a result, our protected model performs quite well, marginally outperforming the (very optimized) scikit-learn model. Using a backpropagation algorithm and stochastic gradient descent, deep Learning develops artificial neural systems with several interconnected layers. There may be hidden layers of neurons in the network that have the tanh, rectification, and max-out hyperparameters. Modern features like momentum training, dropout, active learning rate, rate annealed, and L1 or L2 regularization provide exceptional prediction performance. The worldwide model's parameters are multi-threadedly (asynchronously) trained on the data from that node, and the model-based data is then gradually augmented by model averaging over the entire network. The method is executed on a single-node, direct H2O cluster initiated by the operator. The operation is parallel despite there just being a single node involved. The number of threads may be adjusted in the settings menu under Preferences and General. The optimal number of threads for the system is used automatically. Successful predictions in the healthcare data sets are made using the H2O Deep Learning operator. There will be a classification done since its label is binomial. The Splitting Validation operator creates test and training datasets to evaluate the model. By default, the settings of the Deep Learning activator are used. To put it another way, we'll construct two hidden layers, each containing 50 neurons. The Accuracy measure is computed by linking the annotated Sample Set with a Performer (Binominal Classification) operator. Table 3 displays the Deep Learning Model, the labeled data, and the Performance Vector that resulted from the technique. Conclusions: Deep learning algorithms can be used to design systems that report data on patients and deliver warnings to medical applications or electronic health information if there are changes in the patient's health. These systems could be created using deep Learning. This helps verify that patients get the proper effective care at the proper time for each specific patient. A healthcare decision support system was presented using the Internet of Things and deep learning methods. In the proposed system, we examined the capability of integrating deep learning technology into automatic diagnosis and IoT capabilities for faster message exchange over the Internet. We have selected the suitable Neural Network structure (number of best-hidden layers and activation function classes) to construct the e-health system. In addition, the e-health system relied on data from doctors to understand the Neural Network. In the validation method, the total evaluation of the proposed healthcare system for diagnostics provides dependability under various patient conditions. Based on evaluation and simulation findings, a dual hidden layer of feed-forward NN and its neurons store the tanh function more effectively than other NN. To overcome challenges, this study will integrate artificial intelligence with IoT. This study aims to determine the NN's optimal layer counts and activation function variations.
ABSTRACT
Objective To investigate the risk factors for intraoperative hypotension (IOH) in patients undergoing double plasma molecular adsorption system (DPMAS) artificial liver support therapy. Methods Clinical data were collected from 181 patients (670 cases in total) who underwent DPMAS artificial liver support therapy in Liver Disease Center of The First Affiliated Hospital of University of Science and Technology of China from October 1, 2017 to December 31, 2020, and according to the presence or absence of IOH during DPMAS therapy, they were divided into IOH group with 70 patients and non-IOH group with 111 patients.Clinical indicators were compared between the two groups and their association with IOH was analyzed; prognosis was analyzed at 12 and 24 weeks.The independent samples t -test was used for comparison of normally distributed continuous data between groups, and the Mann-Whitney U test was used for comparison of non-normally distributed continuous data between groups; the chi-square test was used for comparison of categorical data between groups.Univariate and multivariate Logistic regression analyses were used to investigate the risk factors for IOH.The Kaplan-Meier method was used to plot receiver operating characteristic (ROC) curves, and the Z test was used for comparison of the area under the ROC curve (AUC) of independent risk factors. Results The univariate Logistic regression analysis showed that female individuals, individuals aged ≥50 years, and individuals with normal or low body mass index (BMI) tended to have a higher risk of IOH (all P < 0.05), and the multivariate analysis showed that normal or low BMI (odds ratio [ OR ]=3.290, 95% confidence interval [ CI ]: 1.523-7.108, P =0.002) and female sex ( OR =5.146, 95% CI : 2.316-11.432, P < 0.001) were independent risk factor for IOH in patients undergoing DPMAS artificial liver support therapy.The ROC curve analysis of female sex+BMI ≤24 kg/m 2 showed that it had an AUC of 0.639 in predicting IOH ( P =0.002).The patients experiencing IOH had a 12-week survival rate of 55.77%(29/52) and a 24-week survival rate of 50%(26/52), and there were significant differences between the two groups in 12-and 24-week survival rates (12-week: 76.53% vs 55.77%, χ 2 =6.887, P =0.009;24-week: 74.49% vs 50.00%, χ 2 =9.080, P =0.003). Conclusion The risk of hypotension was higher in female patients and that with normal or low BMI during DPMAS artificial liver therapy.Patients with IOH had poor survival prognosis at 24 weeks after DPMAS therapy.
ABSTRACT
Objective:To determine classification and clinical features of morphea.Methods:A retrospective analysis was conducted on epidemiological information about clinical manifestations of and laboratory data from 180 patients with morphea, who visited Zhongshan Hospital, Fudan University from January 2010 to July 2021. Two-independent-sample t test was used to compare the age at onset between genders, and chi-square test to analyze differences in clinical characteristics between different genders and subtypes. Results:Among the 180 patients, 123 were females and 57 were males, with a male-to-female ratio of 1∶2.16. The age at onset of morphea was 28.69 ± 17.97 years for female patients, and 29.90 ± 20.67 years for male patients. Among them, linear morphea was the most common type in this study (68 cases, 37.78%), followed by plaque morphea (63 cases, 35.00%), mixed morphea (28 cases, 15.56%) and deep morphea (21 cases, 11.67%). The disease occurred in all age groups, but the age at onset significantly varied among different clinical subtypes ( F = 5.95, P < 0.001). No significant difference was observed in the age at onset or proportion of clinical subtypes between genders ( F = 0.15, P = 0.696; χ2 =2.88, P = 0.410). Atrophoderma of Pasini and Pierini (APP) was very common (62 cases, 34.44%) in the 180 patients, which mainly manifested as plaques or linear lesions, and 26 out of 45 patients with plaque APP and 11 out of 17 with linear APP were both accompanied by other subtypes of morphea. Among the 75 patients tested for autoantibody profiles, 34 (45.33%) presented with positive results. More diverse types of autoantibodies were found in female patients compared with male patients, and antinuclear antibodies, anti-SSA and anti-SSB antibodies were the most common types. There were various types of comorbidities in female patients, but lichen sclerosus et atrophicus and vitiligo were the most common comorbidities in both genders. Conclusion:High incidence and frequent co-occurrence with other subtypes of APP may be the characteristics of Chinese patients with morphea, and it is recommended to classify morphea into plaque, linear, deep and mixed subtypes.
ABSTRACT
Objective To firstly report 4 cases of dermatomyositis characterized by painful palmar eruptions complicated by fatal rapidly progressive interstitial lung disease (RP-ILD) in China.Methods Four patients with dermatomyositis with painful palmar eruptions complicated by fatal RP-ILD were enrolled from the Department of Dermatology,Zhongshan Hospital,Fudan University between December 2014 and April 2017,and their clinical and pathological features were analyzed.Results Among these patients,3 were female and 1 was male.Their age ranged from 47 to 59 years.Of the 4 patients,3 had no muscular involvement.All of the 4 patients had multiple solid red papules or nodules on the bilateral palms,palmar and lateral surfaces of fingers,which preceded,followed or concurred with the onset of other skin lesions of dermatomyositis.The occurrence of type Ⅰ respiratory failure was preceded by 3 weeks to 5 months of painful palmar eruptions in the 4 patients.Early-stage palmar eruptions were easily misdiagnosed as contact dermatitis,eczema or erythema multiforme.Histopathological examination of the skin lesions on the finger palmar surface showed perivascular infiltration of a few lymphocytes in the dermis,and deposition of varying amounts of mucin-like substances around blood vessels and appendages.Of the 4 patients,3 showed positive staining for anti-melanoma differentiation-associated gene 5 antibody.Although the 4 patients received anti-inflammatory and immunosuppressive therapies,they all finally died of respiratory failure.Conclusions Dermatomyositis with painful palmar eruptions may indicate the occurrence of fatal RP-ILD,and early biopsy of skin lesions is needed to help to identify the disease.Immunosuppressive treatment should be performed timely to improve the prognosis in these patients.
ABSTRACT
Objective To analyze clinical and pathological features of dermatomyositis with panniculitis as a skin manifestation.Methods Clinical data were collected from 9 cases of dermatomyositis with panniculitis as a skin manifestation in Department of Dermatology of Zhongshan Hospital affiliated to Fudan University from October 2012 to July 2016,and their clinical and pathological features were analyzed.Results Of the 9 cases,6 were female and 3 were male,and the age ranged from 28 to 73 years.Panniculitis lesions of the 9 patients all manifested as painful indurated plaques or nodules on the buttock,thigh,waist,back,abdomen,upper extremities and cheeks.These lesions occurred before,after or simultaneously with the onset of characteristic skin and muscle lesions of dermatomyositis,especially preceded the onset of characteristic lesions of dermatomyositis by 30 years in 1 case.Histopathological examination of lesions showed liquefaction degeneration of basal cells,inflammatory infiltration of lymphocytes and plasma cells around blood vessels,in the fat lobules as well as between the lobules and septa in the dermis.The necrosis and calcification of lipocytes,lipomembranous changes,fibrinoid necrosis of damaged vessel walls and microvascular occlusion were observed in some cases.Because panniculitis preceded the onset of characteristic lesions of dermatomyositis,2 patients were misdiagnosed with lupus panniculitis and morphea profunda for several times.Most patients had good response to systemic glucocorticoids combined with immunosuppressive agents,while the patients with lipomembranous fat necrosis had poor response to the combination therapy.Conclusions Panniculitis lesions of dermatomyositis are histologically characteristic,and may do not coincide with the onset of characteristic lesions of dermatomyositis.If panniculitis lesions precede characteristic lesions of dermatomyositis,patients will be easily misdiagnosed.Thus,persistent follow-up visit will be of great importance for the diagnosis.
ABSTRACT
Objective To investigate the risk factors of flare in systemic lupus erythematosus(SLE).Methods In this retrospective cohort study132cases of SLE were followed up for1～5years.The risk factors of SLE flare were analyzed by univariate and multivariate logistic regression with computer software,SPSS.Results Of the132patients,82cases had more than one flare,the total number of flares was115.The incidence of flare was0.22per patient-year.The frequency of flares within5years after diagnosis of SLE was higher than that in more than5years.By univariate logistic regression analysis the age younger than40years at the beginning of the study,age of disease onset younger than32years,stress,depression,overwork,pregnancy,suffering from infectious disease,and quick decreasing of prednisone dosage significantly increased the risk of flare(P
ABSTRACT
Objective To investigate the effects of Chinese drugs for nourishing Yin and clearing the heat on regulation of sex hormones and clinical efficacy patients with SLE.Methods Fifty four active SLE patients with Yin deficiency were randomly divided into 2groups.Thirty patients in tr eatment group were treated with Pill of Wind-weed Phellodendron and Rehman nia(Zhibai Dihuang Wan)and Bolus for Replenishing Vital Ess ence(Da Buyin Wan)combined with western medicine for 2months,while 24patients in control group were treated with western medicine alone.Before and a fter treatment,the serum levels of f ollicle-stimulating hormone(FSH),luteinizing hormone,estradiol,pr ogesterone,testosterone,prolactin were measured by ELISA in 54patien ts,respec-tively.Before treatment the serum l evels of FSH,luteinizing hormone,estradiol,progesterone,testosterone,prolactin were measured in 30active SLEpatien ts without Yin deficiency and 25age-matched healthy controls simultaneously.Results The mean levels of FSH,luteinizing h ormone,estradiol were significant ly higher in patients with Yin de-ficiency than those without Yin deficiency and those in heathy controls(P
