PowerPoint Presentation

[Audio] In this slide, we are discussing a new preventive measure called the Gestational Diabetes Mellitus Detection System, or GDMDS. This system is based on a type of artificial intelligence called deep neural networks, or DNNs. This system has been designed to detect gestational diabetes mellitus, or GDM, with higher accuracy and earlier detection than current methods. The GDMDS takes into account various factors such as changes in the patient's lifestyle and medical history to provide an accurate prediction of GDM. We hope that this system will be a valuable tool in helping to prevent and diagnose GDM more effectively..

[Audio] I will discuss the starting phase of our research, which is the literature review of our study on gestational diabetes mellitus (GDM). Our paper examines how deep neural networks can be used as a proactive measure to detect GDM. GDM can have severe health consequences, so any proactive steps taken to detect it should be promoted. We studied research papers on GDM from different organizations and universities to understand the current state of knowledge. Our analysis revealed that GDM is a complex condition that needs a multi-faceted approach. We used the literature review to plan our research and create an activity diagram for our investigation. We then developed a prototype to test our idea of using deep neural networks and evaluated it through surveys and analysis. We implemented the prototype and tested it against known cases of GDM. Our research found that our approach could help detect GDM in an earlier stage, thus aiding in the prevention of any long-term complications..

[Audio] GDM, otherwise known as gestational diabetes mellitus, is a serious issue impacting pregnant women. Recent research published in SQU Medical Journal revealed the prevalence of GDM to be particularly alarming amongst pregnant Omanis. As a solution, we propose a novel approach which utilizes artificial intelligence and deep neural networks to identify and manage GDM. We are confident that this approach could potentially result in improved outcomes for pregnant women with the condition..

[Audio] We will examine how modern technologies can be utilized to construct a sophisticated preventive measure for recognizing gestational diabetes mellitus (GDM). We will examine Deep Learning with a CNN program, and comprehend and apply Artificial Intelligence. Our problem-solving proficiencies will be increased and enhanced with modern technologies. Finally, we will investigate health-related data on gynaecology and women's health problems..

[Audio] GDM is an increasingly concerning issue for pregnant women in Oman due to its increasing incidence. This condition carries numerous health risks for both mothers and their infants, which emphasizes the need for early diagnosis and treatment. Several risk factors have been linked to GDM, such as obesity, prior gestational diabetes, and advanced maternal age. GDM can also contribute to other conditions, like stillbirth, obesity and heart disease. Traditional treatments often have mixed success, however recent advances in artificial intelligence and machine learning have given researchers better methods to predict and diagnose GDM, such as convolutional neural networks and multilayer perceptrons. This paper discusses these advancements and presents an improved preventative measure, using deep neural networks to detect GDM..

[Audio] In this paper, we look at the need for an advanced preventive measure for the detection of gestational diabetes mellitus (GDM). We explore the functional and system requirements of the new system, and the statement of the problem. Reference is made to Dyck's study which links GDM to maternal obesity and future health issues. The proposed system would provide a chance for women to detect this condition at an early stage, and take the necessary preventive measures to avoid any associated risk. The system would also advise women to pay closer attention to their health..

[Audio] GDM is a common yet potentially dangerous condition for pregnant women and their fetuses. To reduce the risks associated with GDM, early detection is essential. This paper proposes a new approach to prevent GDM using deep neural networks. This system offers a number of advantages, including decreased likelihood of incorrect decisions, and earlier detection of the onset of GDM. Potential risks such as obesity and high blood sugar levels must be taken into account during the management of GDM. Furthermore, the system must possess certain non-functional requirements like scalability, usability, and reliability to be most effective..

[Audio] "Gestational diabetes mellitus (GDM) is a serious health concern for expecting mothers, as it can lead to considerable health risks and at worst even maternal mortality. This paper looks to tackle this problem by introducing advanced preventive measures using deep neural networks. We use various types of multi-layer perceptrons and convolutional neural networks that can help identify GDM in an early stage, so women have an opportunity to take preventive action. The focus of the study is on GDM detection in Oman, given its high rate of pregnancies in the region. With these preventive measures, we can better protect expecting mothers and their children from the potential risks of GDM..

[Audio] Deep neural networks have the potential to facilitate early detection of gestational diabetes mellitus (GDM). This paper examines the use of Google Colab, Python, Keras, PyTorch, TensorFlow, and GRadio in Anaconda to prepare a dataset and apply DNN/FFNN algorithms for prediction. It also looks at the results of the early testing, which reveals a percentage of pregnant women with GDM and can be divided into healthy, GDM, and overt GDM groups. Additionally, it goes on to discuss further management options, including medical nutrition therapy, exercise, and pharmacologic therapy..

[Audio] GDM is a major public health issue for pregnant women and their babies, and research has highlighted the various complications associated with it, such as preterm birth, macrosomia, and higher birth weights. A survey conducted revealed that the majority of people have knowledge of GDM and its causes, with 84.3% aware of what GDM is, 69.4% recognizing that excess sugar consumption can lead to it, and 45.4% indicating that stress is a factor. Furthermore, 95.4% of respondents said that GDM is incurable and 82.4% felt that it can have an impact on their blood pressure. These results demonstrate that people generally comprehend GDM, thus the proposed advanced preventive measure using deep neural networks for detecting GDM could prove to be an invaluable resource for pregnant women and their babies..

[Audio] Examining the dataset models and the age distribution of the participants can provide valuable insights into the critical factors associated with GDM. Family history and the general health of the pregnancy are also essential to assess the risk of GDM. Age has been found to play a significant role in pregnancy outcomes, and should be accounted for when looking at the risk of developing GDM..

[Audio] We see that this slide focuses on the predictive capabilities of deep neural networks for detecting Gestational Diabetes Mellitus. The data presented includes labels, valid results, mean, standard deviation, minimum, and maximum values. Specifically, 3525 valid results were recorded, with the mean value being 0.49 and a standard deviation of 0. Ultimately, the accuracy achieved was 100%, with the minimum and maximum values being 0 and 1, respectively..

[Audio] The results of an analysis of Hemoglobin and Oral Glucose Tolerance Test (OGTT) for diagnosing GDM reveal that Hemoglobin levels range from 18 to 186, with a mean of 8.8 and a standard deviation of 12.7. OGTT range from 48.2 to 195 with a mean of 80 and a standard deviation of 13. This shows that the patients have higher Hemoglobin levels than OGTT on average. These results give us valuable insight into how GDM can be diagnosed more accurately..

[Audio] This slide presents the results of calling the value counts function on the Age field. This gives us a count of the distinct values in the dataset. Calling the plot method with the kind parameter set to bar produces a bar chart displaying a visual representation of the data..

[Audio] We implemented our proposed deep neural networks for detecting GDM in four stages. Initially, we acquired GDM data from multiple sources. Then, we preprocessed the data to ensure the accuracy of the results. Following that, we trained the deep neural networks using the preprocessed data and our proposed model structure. Lastly, we tested our proposed model on the testing dataset to determine accuracy. In this slide, we present the implementation process of our proposed deep neural networks which enables us to effectively detect GDM and provide a more advanced preventive measure..

+ Code + Text + Code + Text Hello [ ] 0k for dirname, filenames in os.walk(' /content/'): for filename in filenames: print(os .path.join(dirname, filename)) File , line 1 0k for dirname, , filenames in os.walk(' /content/'): SyntaxError: invalid syntax SEARCH STACK OVERFLOW.

[ ] def plotPerColumDistribution(df, nGraphShown, nGraphPerR%): nunique : df.nunique() df : for col in df if nunique[col] > 1 and nunique[col] < n]] For displaying purposes, pick columns that have between 1 and unique nRow, nCol : df.shape columnNaæs : list(df) nGraphRou : (nCol + nGraphPerRow • 1) / nGraphPerRou plt.figure(num None, figsize (S * nGraphPerRow, 8 nGraphRow), dpi 88, facecolor for i in nGraphShown)): plt. subplot(nGraphRou, nGraphPerRou, i + 1) columnDf df.iloc[:, i] if (not valueCounts = valueCounts. plot. bar() else: plt.ylabel('counts') plt.xticks(rotation : 9) (column {i)) ') = 1.0, w yad : I.e, hyad : I.e) plt.shou() 'w' , edgecolor -.

[Audio] We propose an innovative technique for detecting Gestational Diabetes Mellitus (GDM). This involves the utilization of Deep Neural Networks and Correlation Matrix Analysis. To begin with, Correlation Matrix Analysis was used with the purpose of narrowing down non-NaN or constant columns and dropping them from the data frame. Thereafter, the correlation between the leftover features was visualized by means of a heatmap. Through this, it is possible to comprehend the connections between the features of the dataset, which can be employed to strengthen a Deep Neural Network model for GDM identification..

[Audio] This research paper aims to develop a preventive measure using deep neural networks for the detection of gestational diabetes mellitus (GDM). To this end, a correlation matrix was created by using the data frame of the analyses and by selecting the columns with more than one unique value. This correlation matrix was then visualized in a color-coded matrix and was used to construct the predictive model for GDM. The proposed method is based on the integration of deep learning and correlation matrix techniques, which allows for a more accurate and reliable prediction of GDM. This predictive model can be used to detect GDM more efficiently and to provide a better healthcare service..

[Audio] We used a phpMyAdmin interface with a LJ Structure SQL Search engine to demonstrate the use of deep neural networks in our GDM research. As you can see, this database has different datasets, including test results. We applied the “filter rows” function to select datasets. After that, we conducted a search, edit, copy or delete a row, and obtained the results. This new approach has the potential to bring about advancements in the prevention and cure of diabetes mellitus, as well as improved accuracy..

[Audio] An advanced preventive measure using deep neural network for detecting gestational diabetes mellitus (GDM) is presented in this paper. This is a critical issue, especially in Oman, where the prevalence of GDM is higher. Our dataset includes a variety of age groups and geographic regions, offering a comprehensive view of GDM prevalence and impact. Notably, GDM affects both mother and fetus, and might continue to have an impact on the child after birth. To reduce the risks, pre-conception lifestyle changes and non-pharmacological approaches such as diet and exercise can be used to keep glucose levels healthy. Medication may also be required, but postpartum screenings should remain in the plan to ensure that women with GDM are closely monitored..

[Audio] We have discussed a new preventive measure for detecting Gestational Diabetes Mellitus (GDM) using deep neural networks. We have seen how this advanced method can help to detect GDM earlier than other techniques, thus allowing the medical team to take the necessary steps for treatment sooner and improve the health and well-being of pregnant women, who are at an increased risk of developing GDM. Any questions?.