Abstract
STUDY QUESTION
Is electronic digital classification/staging of endometriosis by the EQUSUM application more accurate in calculating the scores/stages and is it easier to use compared to non-digital classification?
SUMMARY ANSWER
We developed the first digital visual classification system in endometriosis (EQUSUM). This merges the three currently most frequently used separate endometriosis classification/scoring systems (i.e. revised American Society for Reproductive Medicine (rASRM), Enzian and Endometriosis Fertility Index (EFI)) to allow uniform and adequate classification and registration, which is easy to use. The EQUSUM showed significant improvement in correctly classifying/scoring endometriosis and is more user-friendly compared to non-digital classification.
WHAT IS KNOWN ALREADY
Endometriosis classification is complex and until better classification systems are developed and validated, ideally all women with endometriosis undergoing surgery should have a correct rASRM score and stage, while women with deep endometriosis (DE) should have an Enzian classification and if there is a fertility wish, the EFI score should be calculated.
STUDY DESIGN, SIZE, DURATION
A prospective endometriosis classification proof of concept study under experts in deep endometriosis was conducted. A comparison was made between currently used non-digital classification formats for endometriosis versus a newly developed digital classification application (EQUSUM).
PARTICIPANTS/MATERIALS, SETTING, METHODS
A hypothetical operative endometriosis case was created and summarized in both non-digital and digital form. During European endometriosis expert meetings, 45 DE experts were randomly assigned to the classic group versus the digital group to provide a proper classification of this DE case. Each expert was asked to provide the rASRM score and stage, Enzian and EFI score. Twenty classic forms and 20 digital forms were analysed. Questions about the user-friendliness (system usability scale (SUS) and subjective mental effort questionnaire (SMEQ)) of both systems were collected.
MAIN RESULTS AND THE ROLE OF CHANCE
The rASRM stage was scored completely correctly by 10% of the experts in the classic group compared to 75% in the EQUSUM group (P < 0. 01). The rASRM numerical score was calculated correctly by none of the experts in the classic group compared with 70% in the EQUSUM group (P < 0.01). The Enzian score was correct in 60% of the classic group compared to 90% in the EQUSUM group (P = 0.03). EFI scores were calculated correctly in 25% of the classic group versus 85% in the EQUSUM group (P < 0.01). Finally, the usability measured with the SUS was significantly better in the EQUSUM group compared to the classic group: 80.8 ± 11.4 and 61.3 ± 20.5 (P < 0.01). Also the mental effort measured with the SMEQ was significant lower in the EQUSUM group compared to the classic group: 52.1 ± 18.7 and 71.0 ± 29.1 (P = 0.04). Future research should further develop and confirm these initial findings by conducting similar studies with larger study groups, to limit the possible role of chance.
LIMITATIONS, REASONS FOR CAUTION
These first results are promising, however it is important to note that this is a preliminary result of experts in DE and needs further testing in daily practice with different types (complex and easy) of endometriosis cases and less experienced gynaecologists in endometriosis surgery.
WIDER IMPLICATIONS OF THE FINDINGS
This is the first time that the rASRM, Enzian and EFI are combined in one web-based application to simplify correct and automatic endometriosis classification/scoring and surgical registration through infographics. Collection of standardized data with the EQUSUM could improve endometriosis reporting and increase the uniformity of scientific output. However, this requires a broad implementation.
STUDY FUNDING/COMPETING INTEREST(S)
To launch the EQUSUM application, a one-time financial support was provided by Medtronic to cover the implementation cost. No competing interests were declared.
TRIAL REGISTRATION NUMBER
N/A.
PD58-04 MODELING AUTOMATED ASSESSMENT OF SURGICAL PERFORMANCE UTILIZING COMPUTER VISION: PROOF OF CONCEPT