Status Review:
* Dave Skramsted: Went through the reading notes, using SciLab. Now, I'm building the sensor.
* Larry Beatty: Finished changes to the Sphygmochron application. Germaine found the source for an older least squares Fortran program, using the Watcom and DotNet Fortran complier. Version 2 is the existing program, Version 3, does everything that is shown in the Data Analysis Methods Powerpoint Version 2 doesn't compute fractionated excess, just global excess. Germaine will handle the webpage, describing how to use Sphygmochron version 2. Beyond the original 3 changes, we changed some of the charting, worked around the restriction of contiguous data, and other things that seemed useful and Germaine wanted.
* Chris Adams: Regarding the Sphygmochron prototype that will be posted on the website, we would like to gather requirements: specifically, 1) what do you want, 2) are the inputs to this all of the inputs that they would want, and the outputs all that they would want. (Germaine: a few answers would be that we want data overtop of the Sphygmochron: 1) information about the patient, their symptoms, gender, age. 2) The input file must be prepared, and it would be nice to not have to prepare it. 3) We get data from several hundred subjects, comparison of two drugs, effect of weight loss, dietary intervention; and we wouldn't use this program. We would like common data to be repeated without the data entry person having to enter it.
* Gerry: arranging to work with Dr. Vergil Slee.
* Chris: completed the assessment of privacy requirements. Wants to present them. Will present next time in 30 minutes. Responded to an interesting abstract Franz provided, about the device used in a telemedicine using internet resources. This could lead to additional requirements. We may want it to connect to another device, a 3rd party workstation. We should be prepared for networking the device, leading to privacy issues and security. Our current model is the medical information bus, a multilayered stack. Wants to know the next needs: 1) process engineering - project infrastructure - to leverage the graduate students - 2-3 month jobs, specifying application requirements for device, security requirements, eliciting analysis and workstation requirements for the prototype. So, next week, he would like: 1) 30 minutes to present privacy requirements, 2) future direction, and 3) access to the standards documents.
* Norm Bayne. He has become familiar with web page management and placed his outline of the quality system. Also, he met with a contact with an FDA accelerator, to help customers get through FDA compliance quickly. His needs are to find people to work on the quality system.
* Chris Fuller: talked with a field engineer for piezio film products at Specialty Products and got access to the project lab. Regarding sensor meetings, Chris is available to meet after the Phoenix meeting on Sundays, and Dave can meet weekday evenings and weekend afternoons. Chris is using the Eagle CAD application to do design, so that he isn't tied to a specific board manufacturers product.
* Bob: Attended a conference on diagnostic imaging and software problems. They want to sell their ideas to companies, but they are finding that companies don't want to use open source software because they feel that they will lose proprietary pricing premiums. However, FDA wants open requirements and full disclosure.
* Germaine: Met with Miguel Revilla. He said that Java may
be appropriate because of security compliance, and he brought
his supercomputer Fortran source code. Additionally, the Halberg
Center received more data from the site in India where we had
a problem. The data came in two files because the person was monitoring
for more time than the program would hold (Sph ver 2), and they
couldn't easily combine the two files into one. (Larry said it
might be due to an impeded control Z, used by Dos for End of File.
He said he could modify the Ver 3 application accommodate it.
Gerry Werth - Minitopic on Clinical Information Systems
Entity Codes- Capturing Clinical Detail
When you go from a specific set of clinical data to a class of data, specific detail is lost.
Reference: Partnerships, Health Commons Institute, Fall 2003, pp 7-8.
Entity Codes: Capturing Clinical Detail - Outline
Goals: Capturing exact diagnosis
* Step 1: Input
- FV Dx: Free Vocabulary Diagnosis
* Step 2: Standardization
- FV Code
- CV Dx: Controlled Vocabulary Disagnosis
- Entity Code
* Step 3 : Coding
- Classification: ICD-9-CM, ICD-10-CM, other.
* Glossary
Goals for Diagnosis Entry
Capture Physician's Exact Clinical Diagnosis
* Physician input must be barrier free
- Must accept the physician's own words.
* Must retain the exact language of phyusic8ian
* Must also retain the preferred, standardized term forr the diagnosis.
* Diagnosis must be coded to the appropriate classsific8iations
(e.g. ICD-9-CM)
Step 1: Input
Capture the Original Diagnosis
* Coder finds Free Vocabulary Diagnosis
(FV Dx) in the medical record
Step 2: Standardization
Map to Controlled Vocabulary
* f New
Secretariat Maps FV Dx to Controlled vocabulary (CV)
Step 3: Coding
Code to ICD-9-CM
(e.g. incidences of West Nile are lumped with others. If you give
people an ill structured process, they will structure it but everyone
will structure it differently. Reliability will vary unpredictably.)
* System uses Entity Code:
- IDC-9-CM
- ICD-10-CM
Glossary
* Dx: Diagnosis
* FV: free vocabulary
* FC Dx: original diagnosis verbatim.
* FV Code: An accession number assigned to each Free Vocabulary
Diagnosis.
* FV Database: A list off Free Vocabulary diagns9oses used by
physicians, complied from actual usage, and their codes.
* CV: Controlled Vocabulary: A set of preferred terms used to
standardize terminology within a field or profession - e.g. SNOMED0-CT.
* CV Dx: The standard, preferred term for a given diagnosis.
* Entity Code: Diagnosis Entity
* Code: A permanent, unalterable code assigned to each diagnostic
term in the Controlled Vocabulary.
* Secretariat: A permanent facility which maintains the FV Database
mapped to the CV database and provides coders, in real time, FV
codes and cor3espoinding CV Diagnoses and Entity Codes for new
terminology.
Currently, there are only controlled vocabularies, driven by either the clinicians or the billing agencies.
Originally, ICD was developed by pathologists and usurped by billing agencies. It is an international classification, managed by World Health Organization. In the US, the CM, HC, the generic term is called cross-walking to map one set of codes to another. Almost all are cross-walked to DSM then into HC (healthcare codes). This is a large bureaucratic process.
The current process starts at Step 3. (ICD for diagnosis, and CPT for treatment) so that it gets paid for.
Q&A:
* How has this proposal been accepted? He'll find out next week.
* How do we apply this to ambulatory blood pressure monitors?
At least use the FV.
* What are cardiologists doing with blood pressure?
Bob Schlentz: Data Analysis Methods Review
* Bob has created two sine waves, one with a period of 180
units of time, second with a 90 units time period, summed them,
and applied the Sphygmochron algorithm. He'll use the second as
a measure of noise, and change it from a deterministic source
to a random source and see how the solutions behave. He needs
the Excel techniques for finding a matrix inverse and performing
matrix computations . El will email him the process and an example.
He will continue next week.
This page is maintained by Ellis S Nolley. It was last updated on 25 October 2006.
The author(s) provide this information as a public service, and agree to place any novel and useful inventions disclosed herein into the public domain. They are not aware that this material infringes on the patent, copyright, trademark or trade secret rights of others. However, there is a possibility that such infringement may exist without their knowledge. The user assumes all responsibility for determining if this information infringes on the intellectual property rights of others before applying it to products or services.
Copyright (C) 2006 Ellis S. Nolley. Copying and distribution of this page is permitted in any medium, provided this notice is preserved.
Back to the Meeting Archive Page