The Foundational Pillars Of ALT2MedCoding
ALT2MedCoding uses the best in class processes to code the free form clinical text from the digital documents. Please find some features listed below :
The online Federated medcoding process of ALT2MedCoding follows a standard workflow across the ecosystem, involving both the human coders, seamlessly collaborating from the remote locations, and the algorithmic based central auto coding engine. ALT2MedCoding is a hugely process driven ecosystem aiming at the best outcome in the clinical text processing. It identifies and corrects the noisy clinical text with bad spelling errors in general English and the medical terminologies. Once processed through our ecosystem, the normalized clinical text becomes future-proof, for easy and transparent use, time and again, across the institutions, any day, without any ambiguity.
Foundation of the process is based on full English US/UK dictionary as well as more than 7,00,000 medical terms. Being upgraded continuously to suit the varied needs of a large number of institutions.
Extensive use of MeSH (Medical Subject Headings) dictionary. Identifies and normalizes the medical abbreviations, uses different medical synonyms. Dynamically upgraded to serve the local institutional needs.
Scans and captures the indicators of Diabetes and Cardio Vascular diseases from the loosly described symptoms, lab test reports, past history, clinical assertions in all clinical documents by default without any intervention.
Successfully identifies the relevant clinical chunks from the clinical context, significantly affected by the complex clinical assertion modifiers e.g. negation, uncertainty, hypothetical, conditional experiencer etc.
Can detect and effectively process the proper template headers (CHIEF COMPLAINT,HISTORY OF PRESENT ILLNESS, PAST MEDICAL HISTORY etc. or any variations, if used as a regular institutional practice)/word/phrase/multi-clause sentence boundary in the long multi para/multi-page clinical text.
Successfully captures the "granular concept" whether embedded in 1 single digit number/character or 300+ character long alpha-numeric text with the help of our own powerful and modern NLP based implementation of the SNOMED coding process.
Identifies the relevant SNOMED/ICD-10/LOINC Concept Terms in the Clinical text and embeds the relevant Concept ID's.
Can fully code the clinical text in a few seconds or minutes, depending on the noise level, the structural issues in the original clinical text, and the special features to be addressed. The detection and classification of various clinical assertions and structural issues take a slightly longer duration than the usual. Also available the possibilities of the multiple algorithms to fall back on, for extreme noise or difficult structural issues in the clinical text. Though it takes slightly longer duration, than the usual, rarely a part in the clinical text is missed because of such issues. In our blended system, the specialists can take care of the ambiguous special cases.
The ecosystem always updated within 24 hours of the latest International release. Now, compatible with the July 2021 SNOMED International release.
ALT2MedCoding is modular and extensible. "Can accommodate customers' local practices / needs in days/weeks". Does not need months for the incremental change requests. We thrive on Agile and CI/CD. The foundation and the underlying fundamental principle gives us strength to address a bigger canvas in healthcare than just auto-coding the clinical text.
The full spectrum of SNOMED Semantic TAGS, both ICD-10 short-hand and long hand expressions can be mapped on all the clinical documents, as desired.
Can identify the drug names, form, dosages and medication duration in the clinical text. Can connect to the hospital messaging system for auto-alert.
Can identify the simple and complex references to DATE and TIME, AGE, Gender, Nationality, Ethnic groups, Organisations, Location in the clinical text. It can also be dynamically customized and enhanced.
Can offer the output in the readily pluggable JSON format for EHR, and soon also in the Protocol Buffers format.
Can extract specific independent values for the database and also create an "ALERT" system based on the values and concepts to highlight the SPECIAL CARE needed in the specific case, which also can be connected to the back-end instant messaging engine, if available, for instant communication with all the stake holders. By default every clinical document specially analyzed in real time for any reference to Diabetes and cardio-vascular disease symptoms and communicated through the ALERT system. It evaluates even the values of the regular lab test results.This helps the healthcare professionals in taking the considered decisions.
Proactively pursuing HL7/FHIR and speech interface to offer a well-rounded service in the clinical text processing and EHR. Also the development in non-English International languages are being pursued proactively with new independent/mixed language models. Definite plan for the End-to-end encryption ( being experimented) which would help the regular enterprise customers transact their clinical text in our production pipeline, securely through our next generation thin client, specially being developed for massive federated med coding.
Can accept the input in PDF, DOC, DOCX,TXT, JPG, PNG, TIFF formats. The input documents can be remotely uploaded to the cloud storage systems like Google drive and the final output in DOCX/PDF or structured JSON format can be received, seamlessly, at the same protected drive. We are primed for the "REMOTE WORKING", ready for this unfortunate pandemic as well as the future connected world.
The confidentiality is guaranteed and can also be bound by a legal contract, if needed.
New innovations at any point in the process cycle can be ingested as the process is developed in-house and can easily be customized as per the institutional need.
We are available on email, phone for any further clarification.
Write to firstname.lastname@example.org