Scholarlink

<p>Clinical trial drug analysis plays an essential role in ensuring the safety, efficacy, optimization of new therapeutic drugs prior to market release. The presence of high-dimensional, varied and noisy data in clinical trials makes it challenging to extract significant insights and reliably evaluate drug performance. To address these limitations, there is a critical need for a framework that prioritizes valuable data and quickly validates outcomes allowing for faster and more reliable decision making in drug development. This research proposes a novel model of Entropy based Data Prioritization and Validation in Clinical Trial Drug Analysis using Efficient Predefined Time Adaptive Neural Networks (EBDP-VCD-EPTANN). Initially, data is collected from Pfizer’s Personalized Medicine Initiative dataset. Input data is given to pre- processing using Fast Resampled Iterative Filtering (FRIF) to clean the data from input data. The pre- processed data is fed to Prioritization using Bitterling Fish Optimization Algorithm (BFOA) to focus on most relevant and informative data relevant to clinical trial. The prioritized data is given to Entropy Calculation to measures the uncertainty or diversity in the data. The calculated entropy is fed to Efficient Predefined Time Adaptive Neural Networks (EPTANN) for analyzing drug clinical trial as pass and fail. The proposed EBDP-VCD-EPTANN method is examined using performance metrics like Accuracy, Recall, Precision, F1 Pass and F1 Fail. Finally, the performance of proposed EBDP-VCD-EPTANN approach attains higher accuracy, higher precision and also higher recall when analyzed through existing methods like Machine learning approaches and their applications in drug discovery and design (MLA-ADD), Artificial intelligence to deep learning: machine intelligence approach for drug discovery (AI-DL-DD) and Artificial intelligence to deep learning: machine intelligence approach for drug discovery (AI-DL-DD) respectively.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Entropy based data prioritization and validation in clinical trial drug analysis using efficient predefined time adaptive neural networks

  • 期刊论文

  • 发表日期：

    2026年1月6日

查看全文
  • Niveditha P.S,
  • Saju P. John

摘要

Clinical trial drug analysis plays an essential role in ensuring the safety, efficacy, optimization of new therapeutic drugs prior to market release. The presence of high-dimensional, varied and noisy data in clinical trials makes it challenging to extract significant insights and reliably evaluate drug performance. To address these limitations, there is a critical need for a framework that prioritizes valuable data and quickly validates outcomes allowing for faster and more reliable decision making in drug development. This research proposes a novel model of Entropy based Data Prioritization and Validation in Clinical Trial Drug Analysis using Efficient Predefined Time Adaptive Neural Networks (EBDP-VCD-EPTANN). Initially, data is collected from Pfizer’s Personalized Medicine Initiative dataset. Input data is given to pre- processing using Fast Resampled Iterative Filtering (FRIF) to clean the data from input data. The pre- processed data is fed to Prioritization using Bitterling Fish Optimization Algorithm (BFOA) to focus on most relevant and informative data relevant to clinical trial. The prioritized data is given to Entropy Calculation to measures the uncertainty or diversity in the data. The calculated entropy is fed to Efficient Predefined Time Adaptive Neural Networks (EPTANN) for analyzing drug clinical trial as pass and fail. The proposed EBDP-VCD-EPTANN method is examined using performance metrics like Accuracy, Recall, Precision, F1 Pass and F1 Fail. Finally, the performance of proposed EBDP-VCD-EPTANN approach attains higher accuracy, higher precision and also higher recall when analyzed through existing methods like Machine learning approaches and their applications in drug discovery and design (MLA-ADD), Artificial intelligence to deep learning: machine intelligence approach for drug discovery (AI-DL-DD) and Artificial intelligence to deep learning: machine intelligence approach for drug discovery (AI-DL-DD) respectively.