kenneth-holstad commited on
Commit
4c0e551
·
verified ·
1 Parent(s): 7d69673

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. README.md +22 -0
  2. factors.csv +6 -0
  3. result.jsonl +1 -0
README.md ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags: [mis-752, bonus]
4
+ ---
5
+ # What is something someone new to the healthcare data field might be surprised to learn?
6
+
7
+ **Asked by:** kenneth-holstad · **Answered by:** `nvidia/nemotron-3.5-lightning:free` · **On:** 2026-09-14
8
+
9
+ **The model said.** One thing that often surprises people new to healthcare data is that even in a modern hospital, patient information is typically scattered across a dozen different computer systems—lab results, pharmacy records, physician notes, and billing—none of which were designed to share data seamlessly, so getting a complete picture usually requires manual reconciling or expensive “bridge” software, and without that work the data you have is often incomplete or misleading.
10
+
11
+ **The five factors behind it**, as the model listed them (a rough ranking, not measured data):
12
+
13
+ | Factor | Importance (0 to 100) |
14
+ |---|---|
15
+ | Legacy system fragmentation | 90 |
16
+ | Absent data standards | 85 |
17
+ | High integration costs | 75 |
18
+ | Siloed hospital governance | 70 |
19
+ | Manual reconciliation load | 60 |
20
+
21
+ *One row in `result.jsonl`, the factors in `factors.csv`. The smallest version of every lab in
22
+ MIS 752: ask a model, get a result, turn it into data and a picture, publish it under your own name.*
factors.csv ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ factor,importance
2
+ Legacy system fragmentation,90.0
3
+ Absent data standards,85.0
4
+ High integration costs,75.0
5
+ Siloed hospital governance,70.0
6
+ Manual reconciliation load,60.0
result.jsonl ADDED
@@ -0,0 +1 @@
 
 
1
+ {"student": "kenneth-holstad", "asked_at": "2026-09-14T01:21:43+00:00", "question": "What is something someone new to the healthcare data field might be surprised to learn?", "model": "nvidia/nemotron-3.5-lightning:free", "answer": "One thing that often surprises people new to healthcare data is that even in a modern hospital, patient information is typically scattered across a dozen different computer systems\u2014lab results, pharmacy records, physician notes, and billing\u2014none of which were designed to share data seamlessly, so getting a complete picture usually requires manual reconciling or expensive \u201cbridge\u201d software, and without that work the data you have is often incomplete or misleading.", "factors": [{"factor": "Legacy system fragmentation", "importance": 90.0}, {"factor": "Absent data standards", "importance": 85.0}, {"factor": "High integration costs", "importance": 75.0}, {"factor": "Siloed hospital governance", "importance": 70.0}, {"factor": "Manual reconciliation load", "importance": 60.0}]}