Answer: Amazon Textract + Amazon Bedrock (Titan Text or Claude)

**Explanation:** **Amazon Textract** is an AWS service that extracts text, handwriting, and data from scanned documents. Clinical-trial documents are typically in PDF or scanned formats, so Textract is essential for extracting the text content from these documents. **Amazon Bedrock** (with Titan Text or Claude models) is a fully managed service that provides access to foundation models for text generation and summarization. Once the text is extracted by Textract, Bedrock can be used to summarize the clinical-trial documents and extract key findings. **Why other options are incorrect:** - **Option B (Amazon Rekognition + Amazon Translate)**: Rekognition is for image and video analysis, not document text extraction. Translate is for language translation, not summarization. - **Option C (Amazon Transcribe + Amazon S3)**: Transcribe is for speech-to-text conversion (audio/video), not document text extraction. S3 is just storage. - **Option D (Amazon Polly + Amazon Comprehend)**: Polly is for text-to-speech conversion, not document processing. Comprehend is for natural language processing but doesn't handle document extraction or advanced summarization like foundation models in Bedrock. The combination of Textract for document extraction and Bedrock for summarization and key findings extraction provides the most comprehensive solution for processing clinical-trial documents.

Author: Jin H