Skip to content

MRI-based radiomics and deep learning models for predicting non-local ultra-early recurrence in patients with hepatocellular carcinoma after curative thermal ablation

MRI-based radiomics and deep learning models for predicting non-local ultra-early recurrence in patients with hepatocellular carcinoma after curative thermal ablation

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600128899
Enrollment
Unknown
Registered
2026-07-28
Start date
2026-08-01
Completion date
Unknown
Last updated
2026-08-03

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

Conditions

Hepatocellular carcinoma

Interventions

Gold Standard:Dynamic contrast-enhanced CT or MRI: The typical manifestation is that the lesion is significantly enhanced in the arterial phase, while the enhancement decreases in the portal vein phas
Index test:A risk prediction model for non-local UER based on these independent risk factors

Sponsors

Henan Cancer Hospital
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Patients with primary treatment or recurrent disease after radical therapy and refuse surgical resection; 2. Single tumor with a maximum diameter of <=5.0 cm; 3. Multiple tumors, number <=3, with a maximum diameter of <=3.0 cm; 4. No major vascular invasion, no lymph node or extrahepatic metastasis; 5. Performance status with an Eastern Cooperative Oncology Group score of 0; 6. Child-Pugh Class A; 7. No other malignant tumors; 8. Pre-procedure magnetic resonance imaging (MRI) scan completed within one month; 9. Dynamic enhanced CT or MRI scan confirming complete ablation one month after procedure; 10. Post-procedure follow-up until the first recurrence or for no less than one year.

Exclusion criteria

Exclusion criteria: 1. Lack of complete imaging data, including T2-weighted imaging (T2WI) and dynamic contrast-enhanced imaging. (2) Significant motion artifacts on MRI. (3) Concurrent transcatheter arterial chemoembolization (TACE) (excluding the risk validation set).

Design outcomes

Primary

MeasureTime frame
Specificity;non-local ultra early recurrence;

Secondary

MeasureTime frame
Sensitivity;AUC of ROC;Lesion detection rate;recurrence-free survival;accuracy;Overall survival;negative predictive value;positive predictive value;

Countries

China

Contacts

Public ContactHu Hongtao

Henan Cancer Hospital

hht19761213@163.com+86 13592616373

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Aug 10, 2026