DISSECT Ghent Meeting - Research Workshop

DISSECT Ghent Meeting Part 1

Research Workshop Summary

On a warm September morning, DISSECT COST Action delegates from over fifteen countries gathered at the historic Sint Baafshuis in Ghent, Belgium. Organized and hosted by Prof. Wim Ceelen and Dr. Sarah Cosyns, the two-day gathering for the Research Workshop and Third Management Committee Meeting of DISSECT COST Action CA24158 is the concluding activity for grant period 1. The event combined a strategic meeting with the DISSECT Management Committee and an interactive research training for all the members. The meeting focused on the basics of developing systematic reviews, evidence mapping, critical analysis of bias, and surgical mentorship applicable to different resource settings.

The meeting opened on 24 September with the Management Committee reviewing scientific activities, network development, communication strategies, and resource allocation from the inaugural grant period. Through Short-Term Scientific Missions (STSMs) and Virtual Mobility Grants (VMGs), members had already achieved substantial cross-border knowledge transfer. The broad range of topics covered during the two days showed the action's scope and overarching objective: how we can bridge the gap in cancer surgery training through knowledge exchange between high- and low-resource settings. 

Grant and training school topics included tissue engineering, stem-cell laboratory workflows, and minimally invasive breast surgery, to the clinical integration of digital technologies, complex hepatobiliary robotic surgery, multidisciplinary team discussions, and the foundational role of personal mentorship. 

The financial review focused on directing resources toward training, mobility, and dissemination to a broader audience, including surgeons, policymakers, patient advocates, and innovators. Furthermore, the action celebrated the launch of its live Survey Hub and official web platform, providing a centralized route for news, ongoing surveys, and open resources across all working groups.

The afternoon transitioned into an intensive, four-part Research Workshop designed around the complete sequence of a systematic review—moving from the initial research question and literature search to data extraction, statistical synthesis, and the interpretation of evidence certainty.

In the first workshop session, Nele Pauwels and Muguet Koobasi introduced the foundations of high-quality systematic reviews. They guided participants through structured research questions, protocol development, registration, eligibility criteria, and reproducible search strategies. Attendees examined how to blend free-text terms with controlled vocabulary across databases, navigate peer review using the PRESS checklist, and maintain transparent reporting through PRISMA guidelines. The discussion highlighted common search pitfalls—such as incomplete synonym lists or unverified terms—and emphasized that AI could offer useful terminology suggestions, provided researchers rigorously verify every term.

In the second workshop session, Maria Batrakova demonstrated how AI-assisted tools can support a traceable research workflow from literature search to data extraction and evidence mapping.

Using a practical exercise on ambient AI scribes and clinical documentation time, she showcased seed-paper discovery with Consensus and Semantic Scholar, citation network exploration with ResearchRabbit, and structured data extraction using Elicit and NotebookLM. Crucially, the presentation stressed that verification remains non-negotiable: researchers must pre-define extraction fields, preserve exact outcome units and time points, check AI outputs directly against source texts, and maintain an explicit search trail to avoid over-relying on single citation clusters or generated summaries.

The third workshop session, led by Prof. Wim Ceelen, moved into Meta-Analysis in R, stepping from raw study data to pooled estimates and forest plots. Working through code workbooks using the R meta package, participants explored common-effect and random-effects models for continuous, binary, and time-to-event outcomes. The session paid close attention to statistical heterogeneity, prediction intervals, sensitivity analyses, and funnel-plot interpretation. Rather than treating meta-analysis as a mere calculation, the discussion emphasized evaluating underlying assumptions—asking whether a pooled result meaningfully represents the combined studies and investigating funnel-plot asymmetry beyond simple publication bias.

Closing the first day’s scientific program, Dr. Dara Lundon led the fourth workshop session on risk of bias, certainty of evidence, and the GRADE framework. Working with a teaching dataset comparing adjuvant radiotherapy with an early salvage policy after radical prostatectomy, participants rated certainty of outcomes by outcome. The practical exercise required attendees to justify decisions across risk of bias, inconsistency, indirectness, imprecision, and publication bias while assessing confidence intervals against clinically meaningful thresholds. The discussion explicitly separated the statistical certainty of an effect estimate from the strength of a clinical recommendation, which must account for broader patient values, harms, and real-world circumstances.

🎦 To ensure these valuable insights reach an even wider audience, full recordings of Nele and Muguet's opening lectures, alongside Maria’s, Prof. Wim Ceelen’s, and Dr. Dara Lundon’s workshop sessions from the first day, will soon be available on the official DISSECT YouTube channel.

Following this intense management and scientific agenda, attendees concluded the day with a memorable social program organized by Prof. Wim Ceelen and Dr. Sarah Cosyns, enjoying a picturesque boat tour along the canals of Ghent.

As Day 1 came to a close against the backdrop of Ghent’s historic waterways, the strength of the DISSECT network was clear: bridging gaps in cancer surgery training relies not only on rigorous methodologies and new technologies, but on active, cross-border collaboration. 

Stay tuned for recap highlights from Day 2, and keep an eye on our YouTube channel as we roll out this tool and resources to our wider community!