Saudi Cultural Missions Theses & Dissertations
Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10
Browse
3 results
Search Results
Item Restricted Enhancing Breast Cancer Diagnosis with ResNet50 Models: A Comparative Study of Dropout Regularization and Early Stopping Techniques(University of Exeter, 2024-09-20) Basager, Raghed Tariq Ahmed; Kelson, Mark; Rowland, SarehEarly detection and treatment of breast cancer depend on accurate image analysis. Deep learning models, particularly Convolutional Neural Networks (CNNs), have proven highly effective in automating this critical diagnostic process. While prior studies have explored CNN architectures [1, 2], there is a growing need to understand the role of dropout regularization and fine-tuning strategies in optimizing these models. This research seeks to improve breast cancer diagnosis by evaluating ResNet50 models trained from scratch and fine-tuned, with and without dropout regularization, using both original and augmented datasets. Assumptions and Limitations: This research assumes that the Kaggle Histopathologic Cancer Detection dataset is representative of real-world clinical images. Limitations include dataset diversity and computational resources, which may affect generalization to broader clinical applications. ResNet50 models were trained on the Kaggle Histopathologic Cancer Detection dataset with various configurations of dropout, early stopping, and data augmentation [3–6]. Performance was assessed using accuracy, precision, recall, F1-score, and AUC-ROC metrics [7, 8]. The best-performing model was a ResNet50 trained from scratch without dropout regularization, achieving a validation accuracy of 97.19%, precision of 96.20%, recall of 96.90%, F1-score of 96.55%, and an AUC-ROC of 0.97. Grad-CAM visualizations offered insights into the model’s decision-making process, enhancing interpretability crucial for clinical use [9,10]. Misclassification analysis showed that data augmentation notably improved classification accuracy, particularly by correcting previously misclassified images [11]. These findings highlight that training ResNet50 without dropout, combined with data augmentation, significantly enhances diagnostic accuracy from histopathological images. Original Contributions: This research offers novel insights by demonstrating that a ResNet50 model without dropout regularization, trained from scratch and with advanced data augmentation techniques, can achieve high diagnostic accuracy and interpretability, paving the way for more reliable AI-powered diagnostics.17 0Item Restricted Generative AI for Mitosis Synthesis in Histopathology Images(University of Surrey, 2024-09) Alkhadra, Rahaf; Rai, Taran; Wells, KevinIdentifying mitotic figures has been established as an effective method of fighting cancer at its most vulnerable stage. Traditional methods rely on manual, slow, and invasive detection methods obtained from sectioned tissue samples to acquire histopathological images. Currently, Artificial Intelligence (AI) in oncology has produced a paradigm shift in the fight against cancer, also known as computational oncology. This is heavily reliant on the availability of mitotic figure datasets to train models; however, such datasets are limited in size, type, and may infringe on patient privacy. It is hypothesised that the potential of computational oncology can be realised by synthesising realistic and diverse histopathological datasets using Generative Artificial Intelligence (GenAI). This report demonstrates a comparison of Denoising Probabilistic Diffusion Models (DDPM) and StyleGAN3 in generating synthetic histopathology images, with mitotic figures. The MIDOG++ dataset containing human and canine samples with 7 types of tumours was used to train the models. Quality and similarity of generated and real images was evaluated using as Frechet Inception Distance (FID), Mean Square Error (MSE), Structural Similarity Index (SSIM), and Area Under the Curve (AUC) as a part of Receiver Operating Characteristic (ROC) study were incorporated. Our results suggests that the DDPM model is superior in terms of structural accuracy, however, StyleGAN3 capture the colour scheme better.32 0Item Restricted Anatomic, Histologic and Histomorphometric Analysis of the Acetabular Labrum and its Enthesis(The University of Edinburgh, 2024-04-19) Alomiery, Abdulaziz; Alashkham, Abduelmenem; Hall, Andrew; Gillingwater, TomThe acetabular labrum is a fibrocartilaginous tissue attached to the acetabular rim of the hip joint. It plays a vital role in maintaining both static and dynamic hip joint stability, with multiple associations to pain sensation, proprioception, and various hip joint disorders. Various traumatic and pathological conditions can compromise the structural integrity of the labral body and its attachment, leading to disruptions in hip joint stability. The localisation of the labrum within a large load-bearing joint has led to disparities in its structural and functional characteristics at the cross-sectional level. These disparities are localised within distinct labral zones (i.e., inner and outer zones), which are commonly attributed to the functional adaptation of each zone to mechanical loads. Despite the labrum's critical role in hip joint function, its intricate micro-anatomical differences, both within its main body and at its attachment to bone, have remained largely unexplored in the current research. The primary objective of this project was to provide a detailed investigation of several micro-anatomical aspects of the labrum, encompassing labral innervation, vascularity, attachment features, degrees of degenerative changes, and levels of tears. The aim was placed on understanding the effects of labral morpho-functional zones on the distribution dynamics of these micro-anatomical characteristics. Secondly, this project also assesses the relationship between labral vascularity and innervation and their potential implications on labral healing and degeneration. Human tissue was obtained from 9 embalmed cadavers, comprising a total of 16 hemipelves (10 males and 6 females) with an average age of 80 years. Each hip was divided into 8 distinct regions, resulting in a total of 128 regional segments. The histological investigations produced a total of 742 tissue sections, which were utilised in a variety of histological and immunohistochemical staining techniques. Initially, a systematic zone-specific approach for labral assessment was developed to achieve a thorough understanding of the distribution dynamics of labral histologic and histopathologic characteristics across different zonal territories. The histopathologic evaluation of the labrum revealed a significant increase in the severity of multiple degenerative features, which were predominantly concentrated in the inner labral zone near the articular surface. These degenerative changes encompassed alterations in matrix proteoglycan content, cellularity, collagen organisation, and surface, including the lamellar layer. Immunohistochemical analysis and quantification of labral neuro-vascularity revealed a significantly higher concentration in the outer labral zone, near the joint capsule, with notably reduced neuro-vascularity in the inner zone. The analysis of sensory nerve endings revealed distinct distribution patterns for proprioceptive and nociceptive innervations within the labrum. Various types of sensory corpuscles (including Pacini, Ruffini, Golgi, and unclassified corpuscles), along with free nerve endings (both perivascular and non-perivascular), were significantly more concentrated within the outer labral zone. The histological examination of the labral enthesis structure and morphology unveiled a fibrocartilaginous type of attachment. Notably, the attachment of the labrum to the bone exhibited marked differences in structural morphology between the inner and outer zones. The histopathologic and histomorphometric analysis of the enthesis’s layers, including calcified fibrocartilage region (CFC), tidemark (TM) and cement line (CL), revealed a significantly more developed and compact attachment in the inner zone. In contrast, the outer enthesis displayed a notably weaker anchorage, characterised by less defined entheseal features, a higher frequency of entheseal and cortical bone micro-damage, and a greater incidence of inflammatory and degenerative changes. The labral morpho-functional zones play a significant role in shaping multiple micro-anatomic features of the labrum, affecting the distribution of degeneration processes and labral healing capability. The delineation of these distinct zonal frameworks offers insights into the labrum's functional adaptation to its mechanical environment and a zone-specific vulnerability to injury and degeneration.28 0