Sadia Ruhama | Searcharxiv

arXiv2025

Genetic Profile-Based Drug Sensitivity Prediction in Acute Myeloid Leukemia Patients Using SVR

Acute Myeloid Leukemia (AML) is a highly aggressive blood cancer with low survival rates. Hence, emphasizing the importance of the urgent need for effective treatment modalities. In recent times, the advances in cancer genomics have increased our understanding of AML, as a result, enabling precision oncologists to develop personalized treatment based on individual genetic features and increase the survival rate. However, there is a lack of understanding of how effectively genetic features can be used to predict which drugs are the most suitable for individual-tailored treatment. Therefore, this study explores the potential of Support Vector Regression (SVR) in predicting drug sensitivity of AML patients solely based on their genetic profile. The paper utilized a dataset from Genomics of Drug Sensitivity (GDSC) and developed a precise model that identified the most significant genetic features affecting drug response and achieved promising results with an R-squared score of 0.9523 on the validation set and 0.8928 on the test set.

q-bio.OT↗

arXiv2025

Unmasking Deep Fakes: Leveraging Deep Learning for Video Authenticity Detection

Deepfake videos, produced through advanced artificial intelligence methods now a days, pose a new challenge to the truthfulness of the digital media. As Deepfake becomes more convincing day by day, detecting them requires advanced methods capable of identifying subtle inconsistencies. The primary motivation of this paper is to recognize deepfake videos using deep learning techniques, specifically by using convolutional neural networks. Deep learning excels in pattern recognition, hence, makes it an ideal approach for detecting the intricate manipulations in deepfakes. In this paper, we consider using MTCNN as a face detector and EfficientNet-B5 as encoder model to predict if a video is deepfake or not. We utilize training and evaluation dataset from Kaggle DFDC. The results shows that our deepfake detection model acquired 42.78% log loss, 93.80% AUC and 86.82% F1 score on kaggle's DFDC dataset.

cs.CV↗