AI-Driven Model for Predicting Genomic Alterations and Clinical Outcomes Using H and E Imaging with Omics Data Integration
试验速览
- 阶段
- 不适用
- 状态
- 尚未招募
- 发起方
- 入组人数
- 50,000
- 试验地点
- 1
- 主要终点
- The primary outcomes include AI model accuracy in classifying tumors and predicting genomic alterations, metastasis, and survival.
研究概览
简要总结
This study aims to create an advanced computer model using artificial intelligence (AI) to help doctors better understand and treat cancer. By analyzing tissue samples from 50,000 cancer patients, the model will learn to recognize patterns in microscope images of tumors (called H&E images) and match them with important genetic information.
The goal is to predict how aggressive a cancer is, whether it’s likely to spread, how long a patient might survive, and how well they may respond to certain treatments like chemotherapy or immunotherapy. This could help doctors make faster, more accurate decisions about personalized treatment—without always needing expensive and time-consuming genetic tests.
In short, this AI tool could help bring more precise, faster, and cost-effective cancer care to patients by using information that’s already routinely collected during diagnosis
研究设计
- 研究类型
- Observational
入排标准
- 年龄范围
- 18.00 Year(s) 至 90.00 Year(s)（—）
- 性别
- All
入选标准
- •To be eligible for the study, participants must have a confirmed diagnosis of cancer based on histopathological assessment.
- •They must also have archival FFPE tumor tissue available for multi-omics and AI-based analysis, with a minimum tumor nuclei content of 50% to ensure reliable genomic profiling.
- •In case of prospective recruitment, newly diagnosed patients must provide written informed consent for genomic profiling and AI-assisted predictions.
排除标准
- •Patients will be excluded if their tumor samples are of insufficient quality or quantity for sequencing and AI-based histopathology analysis.
- •Those who have undergone neoadjuvant chemotherapy or radiotherapy before sample collection will be excluded to avoid confounding genomic alterations.
- •Additionally, samples with artifacts, excessive necrosis, or poor resolution in H&E slides will be removed from analysis.
结局指标
主要结局
The primary outcomes include AI model accuracy in classifying tumors and predicting genomic alterations, metastasis, and survival.
时间窗: In the first phase, a retrospective analysis will be performed on 50,000 cancer tissue samples across multiple tumor types, where multi-omics profiling and histopathological imaging will be used to discover novel biomarkers.
次要结局
- Secondary outcomes include assessing the AI model’s ability to predict patient response to chemotherapy, immunotherapy, and targeted therapy. The effectiveness of AI-driven risk stratification will be validated against treatment response rates, progression-free survival, and overall survival.(In the second phase, AI model will be trained using data from these retrospective cohorts, allowing for precise classification of tumor subtypes based on their genomic and histological features. The final phase will involve validation using an independent cohort of patients, where AI-driven predictions will be tested against molecular and clinical outcomes.)
研究者
Dr Ashok Kumar Vaid
Medanta- The Medicity