Areej Alotaibi

Ms. Areej Alotaibi

King Saud University, Riyadh, Ar Riyāḑ, Saudi Arabia
orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 January 2025 in Diagnostics
Verified
Journal Article
Open access
PDF file
AAAreej Alotaibi
Areej Alotaibi
DADuaa alSaeed
Duaa alSaeed
Background/Objectives: Early and accurate diagnosis of skin cancer improves survival rates; however, dermatologists often struggle with lesion detection due to similar pigmentation. Deep learning and transfer learning models have shown promise in diagnosing skin cancers through image processing. Integrating attention mechanisms (AMs) with deep learning has further enhanced the accuracy of medical image classification. While significant progress has been made, further research is needed to improve the detection accuracy. Previous studies have not explored the integration of attention mechanisms with the pre-trained Xception transfer learning model for binary classification of skin cancer. This study aims to investigate the impact of various attention mechanisms on the Xception model’s performance in detecting benign and malignant skin lesions. Methods: We conducted four experiments on the HAM10000 dataset. Three models integrated self-attention (SL), hard attention (HD), and soft attention (SF) mechanisms, while the fourth model used the standard Xception without attention mechanisms. Each mechanism analyzed features from the Xception model uniquely: self-attention examined the input relationships, hard-attention selected elements sparsely, and soft-attention distributed the focus probabilistically. Results: Integrating AMs into the Xception architecture effectively enhanced its performance. The accuracy of the Xception alone was 91.05%. With AMs, the accuracy increased to 94.11% using self-attention, 93.29% with soft attention, and 92.97% with hard attention. Moreover, the proposed models outperformed previous studies in terms of the recall metrics, which are crucial for medical investigations. Conclusions: These findings suggest that AMs can enhance performance in relation to complex medical imaging tasks, potentially supporting earlier diagnosis and improving treatment outcomes.

16 December 2023
Verified
Conference Paper
Open access
RARaghad Alnashwan
Raghad Alnashwan
MMMalak Mashaabi
Malak Mashaabi
AAAreej Alotaibi
Areej Alotaibi
HQHala Qudaih
Hala Qudaih
LALamya Albraheem
Lamya Albraheem
The likelihood of car accidents increases during extreme weather conditions, such as fog, winds, snow, rain, etc. While it may not be possible to prevent all such accidents, their incidence can be reduced by taking proper measures. Therefore, an intelligent accident-avoidance system is necessary to predict the severity of accidents based on weather and road conditions. This research paper suggests three machine learning (ML) methods for an Internet of Things (IoT)-based accident severity prediction system. The methods are Random Forest, LightGBM, and XGBoost.The aim is to predict the severity of car accidents based on various weather features using a machine learning model. However, considering the previous work, we observed that the size of datasets is frequently minimal, and some of the research discusses the influence of the weather on the number of accidents. Therefore, we used the Countrywide Traffic Accident Dataset, which covers 2.8 million vehicle accidents in the United States from 2016 to 2021. In conclusion, our methodology appears to be efficient in predicting the severity of car accidents. Among the three methods, LightGBM achieved the highest prediction accuracy (72%), precision (70%), recall (70%), F1-scores (70%), and area curve (AUC) (0.86) of the receiver operating characteristic (ROC) curve.

26 November 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
AAAreej Alotaibi
Areej Alotaibi
DADuaa alSaeed
Duaa alSaeed
Early and accurate diagnosis of skin cancer improves survival rates; however, dermatologists often struggle with lesion detection due to similar pigmentation. Deep learning and transfer learning models have shown promise in diagnosing skin cancers through image processing. Integrating Attention Mechanisms (AMs) with deep learning have further enhanced the accuracy of medical image classification. While significant progress has been made, further research is needed to im-prove detection accuracy. Previous studies have not explored the integration of attention mechanisms with the pre-trained Xception transfer learning model for binary classification of skin cancer. This study investigates the impact of various attention mechanisms on the Xception model's performance in detecting benign and malignant skin lesions. Using the HAM10000 dermatoscopic image dataset, four experiments were conducted. Three models incorporated self-attention (SL), hard-attention (HD), and soft-attention (SF), respectively, while the fourth model used standard Xception without AMs. Results demonstrated the effectiveness of AMs, with models incorporating self, soft, and hard attention mechanisms achieving accuracies of 94.11%, 93.29%, and 92.97%, respectively, compared to 91.05% for the baseline model, representing a 3% improvement. Both self-attention and soft-attention models outperformed previous studies on recall metrics, which are crucial for medical investigations. These findings suggest that AMs can enhance performance on complex medical imaging tasks, potentially supporting earlier diagnosis and improving treatment outcomes.