Domain knowledge graph issue?

[rollend]

Dear Choi,
I do love your concept of graph-based attention model for healthcare. And I am trying to do some works around it.
First thing first, I do get stuck with the graph generated. I tried to using MIMIC as a demo, but the generated graph shows, well, perhaps very detailed example. One patient that has Cancer ICD9 code was mapped to infectious and parasitic diseases. PID 124, old types 186, new types 50, ICD9: V1011 and 101. Due my limited knowledge, I dont think this may correct, and did I miss anything?
Further let me know if you cant reproduce it.
Secondly, I moved your implementation to Tensorflow, it run successfully, maybe later I will do a pull request for your check the codes.
Looking forward to hearing from you.
Regards,
Shen

[mp2893]

Hi Shen,

Thanks for taking interest in my work.
Could you provide more detail how you arrived at your problem? Most of the things you say don't make sense as of now.

• Did you download CCS multi-level grouper?
• Did you use my script to convert the CCS grouper to matrices?
• Did you download MIMIC-III?
• Did you use my script to convert MIMIC-III into matrices?
• What cancer code was mapped to infectious and parasitic disease?
• Infectious and parasitic diseases is the highest-level concept in ICD9 coding system, why would anything get mapped to that?
• I have no idea what you are talking about regarding PID, old type, new type, V1011 and 101.

It's awesome that you implemented GRAM with TensorFlow. I remember someone asking for TF implementation of GRAM a while back. Please upload the code, and I'll take a look at it.

Best,
Ed

[Al-Dailami]

I think the problem is from here!!

gram/build_trees.py

Line 141 in 5afb17b

codeCount = 0

The process of re-mapping the integers to all medical codes is not clear!!

newTypes[rtypes[icdCode]] = codeCount

repeated for all five levels. and finally the newTypes[rtypes[icdCode]] have the last values of the codeCount variable!!!

can you please correct me and explain this process?

Thanks