SHARE YOUR KNOWLEDGE

& EXPERIENCE!

VISIT

click on ‘2013 Conference’

Abstracts should include:

Author(s) name(s): / Y.Kiran Kumar, Kishan Harwalkar, A.K.Narayan
Author(s) title(s) / Senior Technical Specialist, Senior Domain Specialist, Senior System Architect
Affiliation (Company or organization) / Philips Electronics India Ltd
Bangalore.
Postal address / Manyata Embassy Tech Park, Nagawara, Bangalore -560045
E-mail address / ,,
Telephone: / 9980993684
Abstract: (200-750 words, in English) / Multivendor Algorithm Analysis for Nuclear Medicine Applications
Abstract:
Interoperability is a property of a product or system, whose interfaces are completely understood, to work with other products or systems, present or future, without any restricted access or implementation. Interoperability denotes how far the products from different vendors interact with each other seamlessly1. In this paper we present the work we have done on interoperability by analyzing the problems faced by hospitals and radiologists, when they use the Clinical Applications on standard compliant data generated by different vendors. This research work is mainly focused on application level interoperability and looking at amount of deviation of the algorithm results between various vendors’ workstation/products of Nuclear Medicine.
Interoperability helps customers get the most out of technology, and it also encourages innovation in the industrial sphere. In healthcare, without interoperability, hospitals are forced to turn to large vendors that provide suites of compatible devices but that do not specialize in any one area. Interoperability promotes competition, which in turn encourages innovation and quality. The various levels of interoperability in the healthcare domain as per Wiki.1

The various level of Interoperability is as follows:
  1. Level 0 : No interoperability between products.
  2. Level 1: One to one) communication using common protocol like TCP/IP
  3. Level 2 & Level 3: Syntactic and Semantic Interoperability denotes the PACS/ Workstation connectivity.
  1. Level 4 & 5 – Advanced level of interoperability which deals with the application level interoperability between vendors.
In this paper we start from semantic interoperability by taking DICOM Studies from different vendors and running application on these datasets which are supposed to be Syntactic and semantically level interoperable and check the application behavior based on different types of data sets. The advanced level of interoperability deals with the variation of the results between various vendors, which are not in the clinically acceptable range and also, not able to understand each other’s bookmarks.
Clinical Workflow Background:
In a Nuclear Medicine (NM) department, the technologist performs the Scan and prepares the data and application for theNuclear Medicine Physician. The Nuclear Medicine Physician will review the study and perform post processing and interpretation. The Society of Nuclear Medicine (SNM) has guidelines defined for all the organs/applications. The procedure following in the analysis of multi vendor datasets is as follows:
-The scanned datasets from different vendors are fed to post processing workstation for analysis,
-The dataset is used as input for the relevant application provided by vendor-1 and results are observed, then the same activity is performed in application provided by vendor-2. This is done by an NM Physician (same individual) in both cases to avoid operator variability and errors.
The below figure-1, shows the Dicom Connectivity and the procedure followed in the Nuclear Medicine Department to perform the algorithm results comparison between various Vendors.
Figure 1 : Dicom Connectivity between Two Vendors

Vendor 1: Modality with Workstation Vendor 2: Workstation (System Under Test)
Interoperability Issues and its Clinical Relevance:
Issues found in Level 2 and 3 Interoperability Analysis:
The Query and Retrieving of Multivendor data from PACS also have the problem of importing and storing back to PACS which is more troublesome to the Radiologist, as they are not able to save back the datasets/results/reports.
Issues found in Level 4 and 5 Interoperability Analysis:
The results of the various vendors’ applications parameters are varying between the multivendor workstations, which leads to improper diagnosis by the clinicians, which will have adverse effect on the patient.
The algorithm implementations like Muga, Renal- GFR, ERPF, Depth, by each vendor’s algorithm results are different, that can the change the treatment planning for the patient, which is patient hazard.
The launching of Multivendor Datasets ofRenal, Muga in other vendor’s applications also had issues, i.e., Applications sometimes cannot handle the data of other vendors due to either improper DICOM attributes or different interpretations of the standard.
In the detailed paper, the algorithm analysis with results will be provided for the following Nuclear Medicine applications:
  1. Renal – GFR
  2. Renal- ERPF
  3. Renal-Depth
  4. MUGA
From this study we can conclude, operating at the level of semantic interoperability is not sufficient and product developers should also focus on the advanced level of interoperability.
Keywords: Interoperability, Algorithm, Applications, Cardiology, Neurology, Multivendor, Nuclear Medicine.