AI-assisted analyses of SSR ligand PET/CT in gastroenteropancreatic neuroendocrine tumors.

AI-assisted analyses of SSR ligand PET/CT in gastroenteropancreatic neuroendocrine tumors. - GEP NET AI Study

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

DRKS

Registry ID

DRKS00030550

Enrollment

100

Registered

2023-02-01

Start date

2022-12-08

Completion date

Unknown

Last updated

2025-04-07

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

C26.9

Interventions

Group 1: Patients with histologically confirmed gastroenteropancreatic neuroendocrine tumor.

Eligibility

Sex/Gender

All

Age

18 Years to No maximum

Inclusion criteria

Inclusion criteria: - Histologically confirmed gastroenteropancreatic neuroendocrine tumor. - Performance of a baseline and follow-up SSR PET/CT examination with 68Ga-DOTA-TATE, 68Ga-DOTA-TOC, 68Ga-DOTA-NOC, or 18F-SiFAlin-TATE. - Age > 18 years

Exclusion criteria

Exclusion criteria: - Absence of clinical, pathologic, or radiologic findings. - Performance of SSR hybrid imaging with low-dose computed tomography / native diagnostics.

Design outcomes

Primary

Measure	Time frame
Study Objective #1: SSR PET/CT data collected before and during PRRT will be analyzed using deep learning approaches to predict early treatment response. Outcome: predict clinical treatment response (AUC, PR-AUC).	—

Secondary

Measure	Time frame
Study Objective #2: Develop and evaluate ML algorithms for 1) detection and 2) segmentation of NET primary tumor and metastasis in SSR-PET/CT, and 3) automated determination of SSTR-RADS as prognostic score before PRRT. Outcome: 1) "Intersection over union" (IOU) and other metrics for object detection 2) DICE score between automated and manual segmentation; 3) AUC, PR-AUC if applicable. Study Objective #3: Assess intrinsic heterogeneity of neuroendocrine tumors using radiomics parameters from automatically and manually segmented tumors in SSR hybrid imaging early in PRRT. Outcome: correlation of automatically acquired radiomic parameters versus manually acquired radiomic parameters.	—

Measure

Time frame

Study Objective #2: Develop and evaluate ML algorithms for 1) detection and 2) segmentation of NET primary tumor and metastasis in SSR-PET/CT, and 3) automated determination of SSTR-RADS as prognostic score before PRRT. Outcome: 1) "Intersection over union" (IOU) and other metrics for object detection 2) DICE score between automated and manual segmentation; 3) AUC, PR-AUC if applicable. Study Objective #3: Assess intrinsic heterogeneity of neuroendocrine tumors using radiomics parameters from automatically and manually segmented tumors in SSR hybrid imaging early in PRRT. Outcome: correlation of automatically acquired radiomic parameters versus manually acquired radiomic parameters.

—

Countries

Germany

Contacts

Public ContactClemens Cyran

LMU München

Clemens.Cyran@med.uni-muenchen.de+4989440076642

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: Feb 4, 2026