Trial · NCT07898241
Using an Artificial Intelligence Software Tool (iSeg) During Radiation Treatment Planning for Patients With Lung Tumors
NCT07898241, Not applicable, Not yet recruiting, Northwestern University
3 sourcestrials-2026-11-12-0.5.0
Cite this record
ConductScience Library. Using an Artificial Intelligence Software Tool (iSeg) During Radiation Treatment Planning for Patients With Lung Tumors. Trials, record NCT07898241. Release trials-2026-11-12-0.5.0, 12 Nov 2026. https://conductscience.org/library/trials/NCT07898241
On this page
Trial
As registered on ClinicalTrials.gov, in the graph's capture of the registry.
- Title
- Using an Artificial Intelligence Software Tool (iSeg) During Radiation Treatment Planning for Patients With Lung Tumors
- NCT number
- NCT07898241
- Status
- Not yet recruiting
- Phases
- Not applicable
- Enrollment
- 34 participants, estimated
- Lead sponsor
- Northwestern University
- Conditions
- Lung Carcinoma, Lung Non-Small Cell Carcinoma, Metastatic Malignant Neoplasm in the Lung
- Interventions
- Artificial Intelligence-based Intervention, Best Practice, Computed Tomography, Positron Emission Tomography, Radiology, Treatment Planning, Stereotactic Body Radiation Therapy, Survey Administration
- First posted
- 2026-10-27
- Last updated
- 2026-10-27
Source ClinicalTrials.gov
Registered outcomes9
Each outcome the trial registered, primary first, in ClinicalTrials.gov's words.
| List | Outcome | Time frame | Description |
|---|---|---|---|
| Primary | Determination of lesion segmentation and accuracy | Baseline | The Dice Similarity Coefficient (DSC) and the 95th percentile Hausdorff Distance (HD95) will be calculated for each gross tumor volume (GTV) lesion by comparing physician de novo contours (P) with artificial intelligence (AI) contours (A). A linear mixed-effects model will be used to estimate the mean DSC(P, A\*) across all lesions while adjusting for patient-level clustering. The model will produce an estimated marginal mean DSC, its 95% confidence interval, and the one-sided p-value for the pre-specified hypothesis test of whether the mean DSC exceeds a clinically acceptable threshold. HD95(P, A) will be summarized descriptively using means, medians, and interquartile ranges. Sensitivity analyses will be performed if DSC values deviate substantially from normality; in such cases, per-patient mean DSC values will be computed and tested against the 0.72 threshold using a one-sample t-test or Wilcoxon signed-rank test. |
| Primary | Overall contouring time | Baseline | The primary workflow endpoint examines whether AI assistance reduces overall contouring time relative to de novo physician contouring. For each patient (and each lesion when distinct timing is recorded), the time required for physician de novo contouring (T\_P) and for physician review and modification of AI-generated contours (T\_\[A\^\* \]) will be compared. The primary analysis will use a paired t-test to estimate the mean absolute reduction in time and the corresponding percent reduction. If patients contribute multiple lesions with independently recorded times, a linear mixed-effects model with a patient-level random intercept will be used instead of a simple paired test. This model will estimate the mean difference in contouring time and the associated 95% confidence interval. Will test whether T\_P is greater than Atime, indicating a workflow efficiency gain attributable to AI guidance. Percent time reduction, defined as (T\_P- T\_(A\^\* )) / T\_P × 100%, will be summarized. |
| Secondary | Evaluation of internal target volume (ITV) generation | Baseline | Will evaluate ITV generation to characterize AI performance in motion-inclusive target delineation and to assess the clinical utility of AI-assisted respiratory motion integration. Geometric agreement between physician-derived and AI-assisted ITVs will be assessed using DSC, HD95, centroid displacement, contour volume differences, and voxel-wise edit maps. These analyses will characterize systematic differences in motion envelope definition, identify regions most frequently modified by physicians, and evaluate the impact of respiratory motion on AI-assisted target delineation. |
| Secondary | Comparison of physician evaluation of contours versus AI only contours | Baseline | For GTV, secondary segmentation analyses will include comparisons of physician de novo contours versus AI-only contours (P versus \[vs\] A), AI-only contours versus physician-modified AI contours (A vs A\*), and physician de novo contours versus physician-modified AI contours (P vs A\*). These comparisons will quantify both the magnitude and direction of physician edits and assess the extent to which AI-assisted contours converge toward physician-derived reference contours. For each lesion, DSC and HD95 will be summarized using descriptive statistics, and paired analyses (paired t-tests, non-parametric alternatives or mixed-effects models) will be used to compare contour sets. |
| Secondary | Time to progression (Local control) | At 12 months | Will be estimated using the Kaplan-Meier method and reported as the proportion of patients alive without radiographic evidence of recurrence at the treated lesion. |
| Secondary | Local failure | Up to 12 months | Will be defined using standard institutional radiographic criteria. |
| Secondary | Progression-free survival (PFS) | From completion of stereotactic body radiation therapy until progression, physician-assessed progression, initiation of subsequent anti-cancer therapy, or death from any cause, assessed up to 12 months | Will be evaluated using Kaplan-Meier estimation. The 12-month PFS rate and associated confidence intervals will be reported. While not a powered analysis, exploratory associations between contour quality metrics (e.g., DSC, HD95, magnitude of edits) and PFS may be explored using Cox proportional hazards regression; these analyses will be considered hypothesis-generating. |
| Secondary | Changes in pulmonary function | Before stereotactic body radiation therapy to 12 months post-treatment | Will compare Radiation Therapy Oncology Group Pulmonary Function Test Toxicity Scale scores obtained before stereotactic body radiation therapy and at 12 months post-treatment. Analyses will quantify shifts in toxicity grades, summarize the distribution of functional changes, and describe the incidence of clinically meaningful pulmonary decline. If sample size permits paired statistical testing, ordinal tests or non-parametric alternatives may be used, although primary interpretation will remain descriptive given the limited cohort size. |
| Secondary | Incidence of adverse events | Up to 12 months post-treatment | Toxicity will be summarized according to Common Terminology Criteria for Adverse Events version 6.0, with attention to overall incidence, attribution, severity, timing, and patterns of adverse events. The maximum toxicity grade per patient and the incidence of grade ≥ 3 toxicities will be reported. Tabulation of toxicity profiles will characterize the safety and tolerability of stereotactic body radiation therapy delivered using AI-assisted segmentation. |
Source ClinicalTrials.gov
Outcomes0
Each outcome the trial posted results for, in ClinicalTrials.gov's words.
None in this release.
Results by group0
Each measurement the trial posted, group by group, in the words and numbers it posted.
None in this release.
Participant flow0
How many started, completed or left each group, period by period.
None in this release.
Adverse events0
Each adverse event the trial posted, group by group: the participants affected out of those at risk.
None in this release.
Adverse event totals0
The totals the trial posted for each group.
None in this release.
Papers0
The papers the trial lists on ClinicalTrials.gov.
None in this release.
Grants1
The grants NIH RePORTER links to the trial, newest funded first: NIH's, and some of AHRQ's, CDC's, FDA's and VA's. Each shows its title, principal investigators, organization and funder in its newest year.
| Project | Title | Principal investigators | Organization | Funder | Fiscal years |
|---|---|---|---|---|---|
| P30CA060553 | The Robert H. Lurie Comprehensive Cancer Center | PLATANIAS, LEONIDAS C. | NORTHWESTERN UNIVERSITY | NATIONAL CANCER INSTITUTE | 1993 to 2025 |
Source NIH RePORTER
Trial comparisons0
Comparisons of the trial's two groups, calculated by ConductScience from the counts it posted.
None in this release.
Tests0
The ConductCognition tests this trial's registered outcomes name, as ScienceGraph matched them in trial-tests-2026-11-12-0.2.1.
None in this release.
Data dates
The day each source was copied for this release.
Sources 3
- ClinicalTrials.gov's posted results, through CTTI's AACTFreely available. Acknowledgement requested
- ClinicalTrials.govNo license stated
- NIH RePORTER's clinical studies and project filesPublic domain
Each source's terms
- ClinicalTrials.gov's posted results, through CTTI's AACTClinicalTrials.gov data via Aggregate Analysis of ClinicalTrials.gov (AACT), CTTI. CTTI makes its materials freely available and requests acknowledgement. https://aact.ctti-clinicaltrials.org/documentation/299
- ClinicalTrials.govnone stated
- NIH RePORTER's clinical studies and project filesPublic domain (NIH RePORTER)
ClinicalTrials.gov's data carry an international copyright outside the United States: https://clinicaltrials.gov/about-site/terms-conditions.
For software
Read this trial from software.
The record and each of its lists are JSON, twenty rows a page. No key.
GET https://conductscience.org/api/trials/v1/NCT07898241
GET https://conductscience.org/api/trials/v1/NCT07898241/registered