aws-samples / cloud-experiments Goto Github PK

Develop usage example for People Pathing using Rekognition on stored video.

https://docs.aws.amazon.com/rekognition/latest/dg/persons.html

We may want to use the existing Video Analytics notebook to add the usage example. We may also want to think of a relevant use case contextual to office meeting, training room, or similar setting.

We are exploring a new experiment for developing medical prescription ledger using blockchain on AWS. Medical prescriptions once issued by a doctor to a patient are immutable in nature, require audit of changes, need to be interoperable across healthcare systems, and are a system of record. These traits make them an ideal case for blockchain technology. We also want to explore other AWS services like Comprehend Medical which may help extract structure from prescriptions. Other aspects we want to consider are prescriptions generated by telemedicine apps and how can these be integrated into the system of record. Related documents like x-rays, blood test reports could be part of the ledger. Benefits from this experiment which we are seeking include easy introduction into AWS services which can enable patient electronic health record.

Plan for a reusable python package based on API created across notebooks.

1. We may want to represent each notebook API in a separate file for maintainability
2. Review semantic API naming when all statements come together along with top level class or package name
3. Step 2 may lead to changes in respective notebooks where API is built with usage examples
4. Create a notebook with API usage

Researching open datasets for medical prescriptions. UK National Health Service (NHS) publishes the English Prescribing Dataset with SNOMED code. Each record represents a drug prescribed. Patient level data is not available. Medical conditions being treated are missing. The dataset can be used for determining patterns of drug/medicine prescription by location at postcode level of healthcare practice.

A number of datasets related to drug prescriptions in US, Independent Evaluations of COVID-19 Serological Tests, Drug Labeling, and others. https://open.fda.gov/

MIMIC-III (Medical Information Mart for Intensive Care III) is a large, freely-available database comprising deidentified health-related data associated with over forty thousand patients who stayed in critical care units of the Beth Israel Deaconess Medical Center between 2001 and 2012.

The database includes information such as demographics, vital sign measurements made at the bedside (~1 data point per hour), laboratory test results, procedures, medications, caregiver notes, imaging reports, and mortality (both in and out of hospital).

MIMIC supports a diverse range of analytic studies spanning epidemiology, clinical decision-rule improvement, and electronic tool development. It is notable for three factors:

it is freely available to researchers worldwide
it encompasses a diverse and very large population of ICU patients
it contains high temporal resolution data including lab results, electronic documentation, and bedside monitor trends and waveforms.

The admissions table
The callout table
The caregivers table
The chartevents table
The cptevents table
The d_cpt table
The d_icd_diagnoses table
D_ICD_PROCEDURES
The d_items table
The d_labitems table
The datetimeevents table
The diagnoses_icd table
The drgcodes table
The icustays table
The inputevents_cv table
The inputevents_mv table
The labevents table
The microbiologyevents table
The noteevents table
The outputevents table
The patients table
The prescriptions table
The procedureevents_mv table
The procedures_icd table
The services table
The transfers table

Recently, the MIT Laboratory of Computational Physiology (LCP) started hosting the MIMIC-III dataset on the AWS cloud through the AWS Public Dataset program.

https://mimic.physionet.org/iv/access/cloud/aws/

https://aws.amazon.com/blogs/big-data/perform-biomedical-informatics-without-a-database-using-mimic-iii-data-and-amazon-athena/

Create function for Seaborn Correlation visualization from https://github.com/aws-samples/aws-open-data-analytics-notebooks/tree/master/optimizing-data.