Chapter 1: An Introduction to Artificial Intelligence for Medical Sciences
Chapter goal: This is the initial chapter. Subsequently, it encapsulates the specific context and structure of the book. Then, it states the varying medical specialties central to this book. Likewise, it properly presents independent subsets of artificial intelligence. Besides that, it unveils valuable tools for undertaking exercises; Python programming language, distribution package, and libraries. Afterward, it sufficiently acquaints you with different algorithms, including when to carry them out.Sub-topics: Context of the book. The book's central point. Artificial Intelligence subsets covered in this book. Structure of the book. Tools that this book implements. Python distribution package. Anaconda distribution package. Jupyter Notebook. Python libraries. Encapsulating Artificial Intelligence. Debunking algorithms. Debunking supervised algorithms. Debunking unsupervised algorithms. Debunking Artificial Neural Networks.
Chapter 2: Realizing Patterns in Common Diseases with Neural Networks
Chapter goal: This chapter purportedly contains the application of artificial neural networks in modelling medical data. It properly instigates deep belief networks to model data and predicts whether a patient suffers from an ordinary disease (i.e., pneumonia and diabetes). Equally, it appraises the networks with fundamental metrics to discern the magnitude to which the networks set apart patients who suffer from the disease from those who do not.Sub-topics: Classifying patients' Cardiovascular disease diagnosis outcome data by executing a deepbelief network. Preprocessing the Cardiovascular disease diagnosis outcome data. Debunking deep belief networks.o Designing the deep belief network.o Relu Activation function.o Sigmoid activation function. Training the deep belief network. Outlining the deep belief networks predictions. Considering the deep belief network's performance. Classifying patients' diabetes diagnosis outcome data by executing a deep belief network. Outlining the deep belief networks predictions . Considering the deep belief network's performance. Conclusion.
Chapter 3: A Case for COVID-19 Identifying Hidden States and Simulation Results
Chapter goal: This chapter instigates a set of series analysis methods to uniquely discern patterns in the US COVID-19 confirmed cases. To begin with, the Gaussian Hidden Markov Model inherits the series data, models it and identifies the hidden states, including the means and covariance in those states. Subsequently, the Monte Carlo simulation method replicates US COVID-19 confirmed cases across multiple trials, thus providing us with a rich comprehending of the patternChapter content: Debunking the Hidden Markov Model Descriptive analysis Carrying Out the Gaussian Hidden Markov Modelo Considering the Hidden States in US COVID-19 Confirmed Cases with the GaussianHidden Markov Model Simulating US COVID-19 Confirmed Cases with the Monte Carlo Simulation Methodo US COVID-19 confirmed cases simulation results Conclusion
Chapter 4: Cancer Segmentation with Neural Networks
Chapter goal: This chapter typically exhibits the practical application of computer vision andconvolutional neural networks for breast and skin Cancer realization and segmentation. Equally, it shows an approach to filter medical scans by applying canny, luplican, and sobel filters. It concludes by ascertaining the extent to which the networks accurately differentiate scans of patients with and without Cancer.Chapter content: Debunking Cancer. Debunking Skin Cancer Depicting scans of a patient with Skin Cancer. Classifying Patients' Skin Cancer Diagnosis Image Data by Executing a Convolutional Neural Network.o Preprocessing the training Skin Cancer Image Data.o Preprocessing the Validation Skin Cancer Image Data.o Generating the Training Skin Cancer Diagnosis Image Data.o Tuning the Training Skin Cancer Image Data.o Executing the Convolutional Neural Network to Classify Patients' Skin CancerDiagnosis Image Data.o Considering the Convolutional Neural Network's Performance.o Debunking Breast Cancer. Classifying Ultrasound Scans of Breast Cancer Patients by Executing a Convolutional Neural Network.o Preprocessing the Validation Breast Cancer Image Data .o Preprocessing the Validation Breast Cancer Image Data .o Generating the Training Breast Cancer Diagnosis Image Data.o Tuning the Training Breast Cancer Image Data.o Executing the Convolutional Neural Network to Classify Patients' Breast CancerDiagnosis Image Data.o Considering the Convolutional Neural Network's Performance. Conclusion.
Chapter 5: Modelling Magnetic Resonance Imaging and X-Rays by Carrying out Artificial Neural Networks
Chapter goal: This chapter intimately acquaints you with the practical application of computer vision and artificial neural networks in neurology and radiology. It promptly carries out convolutional neural networks for image classification. The initial network models MRI scans to set apart patients with and without a brain tumor, and the second network models X-ray scans to set apart patients with and without pneumonia. Besides that, it unveils an effective technique for appraising networks in medical image classification.Sub-topics: Debunking Brain Tumors. Classifying Patients' Model Magnetic Resonance Imaging (MRI) Data by Executing aConvolutional Neural Network.o Depicting MRI Scan of Patients with a Brain Tumor.o Depicting Brain Scans without a Brain Tumor.o Preprocessing the Training MRI Image Data.o Preprocessing the Validation MRI Image Data.o Generating the Training MRI Image Data.o Tuning the Training MRI Image Data.o Executing the Convolutional Neural Network to Classify Patients' MRI Image Data.o Considering the Convolutional Neural Network's Performance. Debunking Pneumonia.o Classifying Patients' CT scan Data by Executing a Convolutional Neural Network.o Depicting an X-Ray scan of a Patient with Pneumonia.o Depicting an X-Ray scan of a Patient without Pneumonia.o Processing the X-Ray Image Data.o Generating the Training Chest X-Ray Image Data.o Preprocessing the Validation Chest X-Ray Image Data.o Generating the Validation Chest X-Ray Image Data.o Tuning the Training Chest X-Ray Image Data.o Executing the Convolutional Neural Network to Classify Patients' Chest X-Ray ImageData. Considering the Convolutional Neural Network's Performance. Conclusion.
Chapter 6: A Case for COVID-19 CT Scan Segmentation
Chapter goal: This chapter presents an approach for carrying out convolutional neural networks to model chest CT scan images and differentiate between patients with and without COVID-19.Sub-topics: Classifying Patients' Model Magnetic Resonance Imaging (MRI) Data by Carrying out aConvolutional Neural Network.o Depicting a Chest CT scan of a COVID-19 Negative Patient.o Depicting a CT scan of COVID-19 Negative Patient.o Preprocessing the Training COVID-19 Data.o Preprocessing the Validation COVID-19 CT Scan Data.o Generating the Training COVID-19 CT Scan Data.o Tuning the Training COVID-19 CT Scan Data. Data.o Considering the Convolutional Neural Network's Performance. Conclusion.
Chapter 7 Modelling Clinical Trial DataChapter goal: This chapter familiarizes you with the prime essentials of the most widespread method for adequately investigating data from a clinical trial, recognized as a survival method. It debunks the Nelson-Aalen additive model. To begin with, it encapsulates the method. Subsequently, it promptly presents exploratory analysis, then correlation analysis by carrying out the Pearson correlation method. Following that, it outlines the survival table, then fits the model. It concludes by carefully outlining the profile table, confidence interval, and reproducing the cumulative and baseline hazard.sub-topics: Debunking Clinical Trials. An Overview of Survival Analysis. Context of the Chapter. Exploring the Nelson-Aalen Additive Model. Descriptive Analysis. Realizing a Correlation Relationship. Outlining the Survival Table. Carrying out the Nelson-Aalen Additive Model.o Outlining the Nelson-Aalen additive Model's Confidence Intervalo Discerning the Survival Hazard.o Discerning the Cumulative Survival Hazard.o Baseline Survival Hazard. Conclusion. References.
Chapter 8: Medical Record Categorization
Chapter goal: This chapter sufficiently apprises a wholesome approach for realizing patterns in medical records by carrying out a linear discriminant analysis model. To begin with, it summarizes medical recording. Subsequently, it exhibits a technique of cleansing textual data by carrying out fundamental methods like regularization and TfidfVectorizer. Afterward, it executes the method to classify the medical specialty, then it assesses the extent to which it segregates classes.Sub-topics: Medical Records. Context of Chapter. Debunking Categorization with Linear Discriminant Analysis.o Descriptive Statistics.o Preprocessing the Medical Records Data.o Carrying out Regular Expression.o Carrying Out Word Vectorization.o Carrying out the Linear Discriminant Analysis Model to Classify Patients' MedicalRecords.o Considering the Linear Discriminant Analysis Model's Performance. Conclusion.
Chapter 9: A Case for Psychology: Factoring and Clustering Personality Dimensions
Chapter goal: This chapter introduces you to analyzing the underlying patterns in human behavior by promptly carrying out exploratory factor analysis and cluster analysis. To begin with, it covers the big five personality dimensions. Following that, it presents an approach for typically collecting data by retaining a Likert scale and measuring the reliability of the scale with Cronbach's reliability testing strategy. Subsequently, it performs factor analysis; beginning with estimating Bartlett Sphericity statistics, then the Kaiser-Meyer-Olkin statistic. Following that, it rotates the eigenvalues by carrying out the varimax rotation method and estimates the proportional variances and cumulative variances. In addition, it executes the K-Means method to observe clusters in the data; beginning with standardizing the data and carrying out principal component analysis.Sub-topics: Debunking Personality Dimensions. Questionnaires. Likert Scale. Reliability.o Spearman-Brown Reliability Testing Strategy.o Carrying out Cronbach's Reliability Testing Strategy. Carrying out Factor Model.o Carrying out the Bartlett Sphericity Test.o Carrying out the Kaiser-Meyer-Olkin Test.o Discerning K with a Scree Plot.o Carrying out Eigenvalue Rotation. Varimax Rotation. Carrying out Cluster Analysis.o Carrying out Principal Component Analysis.O Returning K-Means label.
