IDENTIFY HIGH RISK HEALTH RECORDS

Overview
Oktank is a hospital provider who is figuring out the challenge of high operational cost. One of the goals to reduce cost is to automate handling of medical records.

CEO has directed IT team to create a technical solution to read medical records . One of the requirement is to make it easier for doctors or nurses to quickly decipher medical documents and identify high risk patients.

Possible Points: 150 Clue Penalty: 0 Points Earned: 150
Completed!
Background Section:

Oktank has a process that scans medical documents that are provided by specialist doctors and saves it as text files stored in Amazon S3 bucket.

CTO of oktank has instructed the development team to use comprehend medical to identify relevant medical data from these documents.

Your Task Section:

As a part of development effort, you are required to

Update a lambda function “ComprehendMedicalProcessingFunction” and use comprehend medical detect_entities_v2 api to read a test medical document file “patient_medical_document.txt” that belongs to patient ABC . This file is stored in the S3 bucket.
This function should read the response from comprehend medical to identify high risk Medical condition “Metastatic breast cancer” and return the confidence score of this identification. This confidence score can be used by generalists doctors to identify if the patients suffers from “Metastatic breast cancer”
Function should return the trait score value in the following format {‘status’ : 200, ‘body’: json.dumps(confidence_score_value_here)}
GETTING STARTED:

Checkout the Output properties tab for resource details

TASK VALIDATION:

The task will be validated automatically or you can click “Check Progress”.

Inventory Section:

AWS Lambda Comprehend Medical S3

Services You Should Use:

AWS Lambda Comprehend Medical

需要修改一点点，detectOutputRow里有许多Score，我们需要返回最高值。