Storage of medical records etc

Olinda99

the folders need to be named yyyymmdd eg 19801203

or, if you just want a folder by month then yyyymm eg 198012

400ixl

Chuck all the docs in a folder, point Google NotebookLM at the folder and ask whatever questions you want of the data in natural language.

Use AI as the tool rather than trying to pull manual structure around unstructured documents.

Olinda99

good point - I guess AI is the way forward !

PHK

MouldyOldDough said:

Next problem - how do I sort the list of 13248 folders into a more user friendly order ?

ie) Actual date order - from 01 Jan 89 to 31 Dec 25

At the moment - it's alphabetical which means tons of work !

The option doesn't appear to exist

The idea was to have a folder for the year, nested inside that a folder for the month and nested inside that a folder for the days. Use the YYYY format for years, MM for months and DD for days. 

But this is a very time-consuming way to find documents. It would be better to add them to a database that will search inside documents (including scans, photos) and produce indices/indexes in whatever way you want eg by day or doctor or incident etc etc. It won't be free but not too costly. 

B0bbyEwing

RumRat said:

You should be able to access all medical records ant tests etc via the NHS app.

I'm only scanning through the thread but you said "all" and "should". 

Not so. 

Take this from someone who has been trying to access records to bring a negligence claim against the NHS.

I had to get in touch with a department of the NHS, not the app. Even then only some records could be obtained. 

And then one had key information completely blacked out. 

None of it could be obtained via the app. 

Olinda99

my advice would be to have a folder for each year and name the documents within them eg

19880506 - xray scan.pdf

poppellerant

I've just thought, why not name the files like below (in YYYY-MM--DD-##.xxx format) - no need for folders:

• 2000-01-01--01.jpg
• 2000-01-01--02.png
• 2000-01-01--03.docx
• 2000-01-02--01.jpg
• 2000-01-02--02.png
• 2000-01-03--01.docx

That's the general idea.

You could build further on that and put brief descriptions in brackets, for example:

• 2000-01-01--01 (X-ray of left ankle).jpg

400ixl

We aren't in the 90's. No need for any of this messing around with renaming files or databases (which are for structured data non unstructured documents). Use Ai for what it was designed for, let it RAG the documents and index the into a knowledge base.

poppellerant

400ixl said:

We aren't in the 90's. No need for any of this messing around with renaming files or databases (which are for structured data non unstructured documents). Use Ai for what it was designed for, let it RAG the documents and index the into a knowledge base.

I'd never trust AI.  So, being as polite as possible, you can stick AI.

mksysb

MouldyOldDough said:

I am trying to put my historic medical records onto my google calendar - to make them accessible but it is extremely slow to get several hundred docs on the correct day !

I have put all of the documents (pdf's) on to google drive and pointed google calendar at these PDFs

I does allow unlimited attachments at each day but the problem is that there is no indication that there are any docs attached to a day - unless you open EVERY day - this is not exactly user friendly and also there is NO search facility to search 35 years worth of days for a specific phrase or even word used in the title of any document

So its back to the drawing board

I believe that I should be using access to create a database - but, nowadays my poor memory doesn't know enough to even begin - my days of using access ended with '97 !

A database would be good, but not Access.  Look for a nosql one like MongoDB or CouchDB.