Renal mass, benign/malignant
Conditions
Interventions
Sponsors
Eligibility
Inclusion criteria
Inclusion criteria: 1. Age >= 18 years; 2. Undergoing minimally invasive renal surgery at a participating center, including partial nephrectomy, radical nephrectomy, radical nephroureterectomy, inferior vena cava tumor thrombus/cancer thrombus thrombectomy, pyeloplasty, or other related procedures; 3. Availability of intraoperative video recordings suitable for research, with video quality generally sufficient for anatomical identification; 4. Relatively complete perioperative clinical data, sufficient to support the recording of baseline case characteristics and study outcomes; 5. For the prospective validation phase: cases must meet the requirements for consecutive enrollment and comply with the ethical requirements of each participating center regarding informed consent or waiver of informed consent.
Exclusion criteria
Exclusion criteria: 1. Emergency or critically ill cases requiring immediate surgery and deemed unsuitable for inclusion in the study workflow; 2. Video recordings or medical records that cannot be effectively de-identified; 3. Severe video incompleteness, missing key surgical phases, or excessively low resolution that precludes valid analysis; 4. Explicit refusal during the study period to allow the use of their data for research purposes; 5. Other conditions that, in the investigator’s judgment, make the case unsuitable for inclusion in the study.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Dice coefficient;Intersection over Union, IoU;Precision;Recall;F1 score;Mean inference latency;Processing frame rate, FPS;System crash rate;Case coverage rate;Successful real-time display rate; | — |
Secondary
| Measure | Time frame |
|---|---|
| Class-level performance for each anatomical category;Stratified performance across different surgical procedures;Improvement in anatomical awareness;Likert scale score;NASA-TLX score;Accuracy in human–AI comparison;Sensitivity in human–AI comparison;Specificity in human–AI comparison;Mean annotation time in human–AI comparison; | — |
Countries
China
Contacts
Chinese PLA General Hospital