Methods and system of electronic image analysis
A machine translation of a document is created via a compilation of services by mapping textual content from an image to create a plurality of mapped locations correspondent to at least one object from the image, populating each of the mapped locations with at least one character indicative of the object, each character sharing at least one similar attribute, adding to the image the populated mapped locations, and highlighting at least a portion of the textual content in accordance with the populated at least one character. A compilation of services is provided for identifying, extracting, and assessing electronic images by using a layered approach that reduces time and improves reviewing of medical records and other kinds of documentation.
1. A method for transforming documentation in a computing device, the method comprising:
receiving, by the computing device, a request from a user to transform a document; storing, by the computing device, an original copy of the document and documentation details associated with the document to a database;
extracting, using a Web service, textual content from each page of the document; identifying information about the extracted textual content from each page of the document;
creating an invisible and searchable layer of each page the document based on the extracted textual content;
combining, for each page of the document, an image of the original copy of the document with the corresponding invisible and searchable layer of the document to create a searchable version of the document;
storing the searchable version of the document and the identified information to the database for subsequent retrieval; and
wherein, for each word of the extracted textual content and each page of the document, creating the invisible and searchable layer further comprises:
determining original bounding box size and bounding box coordinates for each word of the extracted textual content based on the identified information;
setting the page transparency of the page to maximum;
reducing, gradually, the font size until the word fits within the original bounding box size; and
writing the word to the page based on the bounding box coordinates of the page.
2. The method of claim 1 , wherein, for each word of the extracted textual content and each page of the document, creating the invisible and searchable layer further comprises:
determining original bounding box size and bounding box coordinates for each word of the extracted textual content based on the identified information;
setting the page transparency of the page to maximum;
reducing, gradually, the font size until the word fits within the original bounding box size; and
writing the word to the page based on the bounding box coordinates of the page.
3. The method of claim 1 , wherein the source document includes one or more of electronic medical records, diagnostic and procedure codes, billing codes.
4. The method of claim 1 , further comprising:
identifying one or more custom terms or phrases in the searchable version of the document;
wherein the one or more custom terms or phrases are included in the request from the user.
5. The method of claim 1 , further comprising: identifying one or more custom terms or phrases in the searchable version of the document; wherein the one or more custom terms or phrases are included in the request from the user.
6. The method of claim 1 , wherein the document includes one or more electronic medical records, diagnostic and procedure codes, and billing codes.
7. The method of claim 1 , wherein the extracting and identifying information from the document uses artificial intelligence, machine learning, or both.
8. The method of claim 1 , wherein the request is received via a web portal, an application programming interface (API}, or a cloud-based file service.
9. The method of claim 1 , wherein the textual content includes printed text and handwritten content.
10. The method of claim 1 , wherein creating the invisible and searchable layer of each page of the document further comprises:
setting a page transparency to maximum.
11. The method of claim 1 , wherein the database is preloaded with one or more datasets.
12. The method of claim 1 , wherein the one or more datasets include interface control documents (ICD}, medications, medical devices, and billing codes.
13. The method of claim 1 , further comprising: identifying one or more pages of the document that include low-quality text or-low-quality handwriting.
14. A system for transforming documentation in a computing device, the system comprising:
a graphical user interface capable of receiving a request from a user to transform a document stored on a non-transitory computer readable storage medium;
at least one network port for receiving a source document and at least one document attribute;
a database for recording the extraction of textual content from at least one page of the source document;
a rules engine for identifying information about the extracted textual content and creating a hidden and searchable layer on at least one page of the source document based on the extracted textual content;
a data layer for receiving a searchable version of the source document and the identified information wherein an image of the source document with the corresponding hidden and searchable layer of the document are merged to create a searchable version of the documents and
wherein the creating the hidden and searchable layer further comprises: determining original bounding box size and bounding box coordinates for each word of the extracted textual content based on the identified information;
setting the page transparency of the page to maximum;
reducing, gradually, the font size until the word fits within the original bounding box size; and
writing the word to the page based on the bounding box coordinates of the page.
15. The system of claim 14 , wherein the source document includes one or more of electronic medical records, diagnostic and procedure codes, billing codes.