Aneeq Zia is the first-listed author of the 2025 Surgical Visual Understanding, or SurgVU, dataset paper. The resource brings together video and labels from robotic surgery training to support machine-learning research. For a field that depends on access to specialized examples, a shared dataset gives researchers a common starting point for developing and comparing ideas.

A Georgia Tech Foundation
Zia’s academic background includes a 2018 Ph.D. in electrical and computer engineering from the Georgia Institute of Technology. His later research record includes collaborative work on surgical datasets and evaluation challenges, extending the scope beyond a single model or experiment.

The SurgVU Dataset
The SurgVU paper lists Zia’s affiliation as Intuitive Surgical. It describes 280 video clips from 155 training sessions, totaling more than 840 hours. Surgeons performed standardized exercises on a porcine model using the da Vinci robotic system. This is training data collected in a controlled setting, an important detail when interpreting its possible uses.

The collection includes tool-presence labels drawn from system information and task annotations supplied by clinical experts. The authors identify challenges including uneven representation of tools and labels that may indicate an installed instrument even when it is not visible. Those details give researchers practical questions to investigate.

A dataset and a model serve different purposes. The dataset supplies examples and annotations; the model is one method of learning from them. Making those inputs available gives other researchers a basis for evaluating approaches, exploring errors and developing alternative methods.

Testing Skills Assessment In Simulation
Zia is also first-listed on the paper reporting the MICCAI 2021 SimSurgSkill challenge. Participants worked with virtual-reality surgical tasks to locate instruments and predict skill. The report described the winning approaches and made the dataset available for further research.

The challenge addressed the difficulty of supplying timely expert feedback during surgical training. Its results were presented as a starting point for future development, rather than proof that automated assessment could already replace expert judgment.

Aneeq Zia, associated with Hamilton, New Jersey, contributes to this shared research effort through his coauthored publications. A dataset leaves room for the next team’s ideas, making it a contribution whose usefulness can extend beyond the experiment that first motivated it.

Sources
1. Georgia Tech advisor alumni directory

2. Surgical Visual Understanding Dataset January 2025

3. MICCAI 2021 SimSurgSkill challenge report