Updated: Sep 1, 2021
Solution Kit The healthcare system is genuinely stretched with a growing population, expanding life expectancy, and now further with the pandemic. The health-tech industry has delivered multiple innovations over the years. With maturing AI and investments in the healthcare space, Ambient Clinical Intelligence has emerged as a critical focus area in heath-tech.
Ambient Clinical Intelligence encompasses leveraging AI across the provider lifecycle and seamlessly integrating with Electronic Health Records systems. The most impactful areas are automated clinical documentation, improving patient data for downstream provider and payer processes, reducing provider workload, and improving the time they can spend with patient care activities. Additional use cases across telehealth, radiology, medical events prediction, payer adjudication are possible with Ambient Clinical Intelligence.
To help jumpstart your Ambient Clinical Intelligence applications, we have assembled a diverse set of software components.
If you are further interested, search to discover more exciting components to jumpstart your application development on kandi.
Speech to Text and NLP
stanza by stanfordnlp, Medico by pranayjoshi, PrescAI by 090max are components that help you experiment on speech to text and NLP with specialization on healthcare.
stanzaby stanfordnlp
6673 Version:v1.5.0Official Stanford NLP Python Library for Many Human Languages
stanzaby stanfordnlp
Python 6673 Version:v1.5.0 License: Others (Non-SPDX)
Cloud Speech to Text
If you would like to compare the speech to text capabilities of the hyperscale cloud providers, try Cloud Speech-to-Text API by Google, Train-Custom-Speech-Model by IBM, and amazon-comprehend-medical-fhir-integrationby aws-samples.
Cloud Speech-to-Text APIby Google
Cloud API 0 Version:Current
Speech recognition
Cloud Speech-to-Text APIby Google
Cloud API 0 Version:Current License: Others (SaaS)
48 Version:CurrentCreate a custom Watson Speech to Text model using specialized domain data
Train-Custom-Speech-Modelby IBM
JavaScript 48 Version:Current License: Permissive (Apache-2.0)
amazon-comprehend-medical-fhir-integrationby aws-samples
35 Version:CurrentWorkshop demonstrating the use of Amazon Comprehend Medical to extract clinical entities from unstructured text and map them to FHIR resources.
amazon-comprehend-medical-fhir-integrationby aws-samples
Java 35 Version:Current License: Others (Non-SPDX)
Automated Clinical Actions
If you further want to experiment on medical event prediction and automated clinical actions, try ehr-rwe by som-shahlab, SequentialPhenotypePredictor by wael34218, MetaPred by sheryl-ai, CogStack-SemEHR by CogStack, clicr by clips.
ehr-rweby som-shahlab
23 Version:CurrentWeak supervision methods for extracting real world evidence from EHRs
ehr-rweby som-shahlab
Python 23 Version:Current License: Permissive (Apache-2.0)
SequentialPhenotypePredictorby wael34218
11 Version:Current
License: No License (null)Early diagnosis predictor using contextual representation of medical events
SequentialPhenotypePredictorby wael34218
Python 11 Version:Current License: No License
MetaPredby sheryl-ai
31 Version:Current License: No License (null)MetaPred: Meta-Learning for Clinical Risk Prediction with Limited Patient Electronic Health Records (KDD 2019)
MetaPredby sheryl-ai
Python 31 Version:Current License: No License
CogStack-SemEHRby CogStack
62 Version:core.0.1Surfacing Semantic Data from Clinical Notes in Electronic Health Records for Tailored Care, Trial Recruitment and Clinical Research
CogStack-SemEHRby CogStack
JavaScript 62 Version:core.0.1 License: Permissive (Apache-2.0)
clicrby clips
127 Version:Current License: No License (null)Machine reading comprehension on clinical case reports
clicrby clips
Python 127 Version:Current License: No License
Patient Data and EHR
To support these applications and experiments, you would need an EHR/ EMR to connect to, such as openemr by openemr, and tons of patient data that you can synthesize from components such as syntheaby synthetichealth.
openemrby openemr
2276 Version:CurrentThe most popular open source electronic health records and medical practice management solution.
openemrby openemr
PHP 2276 Version:Current License: Strong Copyleft (GPL-3.0)
syntheaby synthetichealth
1730 Version:master-branch-latestSynthetic Patient Population Simulator
syntheaby synthetichealth
Java 1730 Version:master-branch-latest License: Permissive (Apache-2.0)