Ashwin Makwana

The limited availability of diverse, high-quality medical images constitutes a significant obstacle to training reliable deep-learning models that can be used in clinical settings. The traditional methods used for data augmentation generate inadequate medical photos that result in poor model performance and a low rate of successful generalization. This research studies the effectiveness of DCGAN cGAN CycleGAN and SRGAN GAN architectures through performance testing in five essential medical imaging datasets, including Diabetic Retinopathy, Pneumonia and Brain Tumor and Skin Cancer and Leukemia. The main achievement of this research was to perform an extensive evaluation of these GAN models through three key metrics: generation results, training loss metrics, and computational resource utilization. DCGAN generated stable high-quality synthetic images, whereas its generator produced losses from 0.59 (Pneumonia) to 6.24 (Skin Cancer), and its discriminator output losses between 0.29 and 6.25. CycleGAN showed the best convergence potential for Diabetic Retinopathy with generator and discriminator losses of 2.403 and 2.02 and Leukemia with losses at 3.325 and 3.129. The SRGAN network produced high-definition images at a generator loss of 6.253 and discriminator loss of 6.119 for the Skin Cancer dataset. Still, it failed to maintain crucial medical characteristics in grayscale images. GCN exhibited stable performance across all loss metrics and datasets. The DCGAN model required the lowest computing resources for 4 to 7 hours, using 0.9M and 1.4M parameters. The framework of SRGAN consumed between 7 and 10 hours and needed 1.7M to 2.3M parameters for its operation, and CycleGAN required identical computational resources. DCGAN was determined as the ideal model for synthetic medical image generation since it presented an optimal combination of quality output and resource efficiency. The research indicates that using DCGAN-generated images to increase medical datasets serves as a solution for boosting AI-based diagnostic system capabilities within healthcare.

Background Achieving optimal tacrolimus dosing is vital for effectively balancing therapeutic efficacy and safety, as CYP3A5 genetic variants and inter-patient variability emphasize the need for precision strategies. Aim This study aimed to optimize tacrolimus dosage prediction for renal transplant recipients by incorporating genetic polymorphisms, specifically profiling CYP3A5 genetic variants, within the DoseOptimal framework to enhance interpretability and accuracy of dosing decisions. Method The dataset comprised clinical, demographic, and CYP3A5 genetic variants information from 1045 stable tacrolimus-treated patients. The DoseOptimal framework was developed by integrating the strengths of the most effective algorithms from fifteen machine learning models. SHapley Additive exPlanations (SHAP) and decision tree insights were incorporated to enhance the framework’s interpretability. The framework’s performance was assessed using mean absolute error (MAE) and the coefficient of determination (R² score). The F-statistic and p value were calculated to validate the framework’s statistical significance. Results The DoseOptimal framework demonstrated robust performance with an R² score of 0.884 in the training set and 0.830 in the testing set. The MAE was 0.40 mg/day (95% CI 0.38–0.43) in the training set and 0.41 mg/day (95% CI 0.38–0.45) in the testing set. The framework predicted the ideal tacrolimus dosage in 87.6% (n = 275) of the test cohort, with 3.2% (n = 10) underestimation and 9.2% (n = 29) overestimation. The framework’s statistical significance was confirmed with an F-statistic of 266.095 and a p value < 0.001. Conclusion The framework provides precision medicine-based dosing solutions tailored to individual genetic profiles, minimizing dosing errors and enhancing patient outcomes.