Minimally invasive neurosurgery using an AI-guided robot for moderate basal ganglia hemorrhage

A Multicenter, RAndomlzed, coNtrolled, umBrella Trial fOr Minimally Invasive Neurosurgery With Al-assisted Robotic Guidance for Moderate Basal Ganglia Hemorrhage (RAINBOW-MBH)

This study will try using an AI-guided robot during minimally invasive hematoma removal to see if people with moderate basal ganglia hemorrhage recover better than with standard care.

Quick facts

• Phase: Not applicable
• Study type: Interventional
• Enrollment: 330 (estimated)
• Ages: 18 Years and up
• Sex: All
• Sponsor: Second Affiliated Hospital, School of Medicine, Zhejiang University (Academic / other)
• Locations: 1 site (Hangzhou, Zhejiang)
• Trial ID: NCT06465719 on ClinicalTrials.gov

What this trial studies

This is a nationwide, prospective, multicenter randomized controlled trial comparing minimally invasive hematoma evacuation guided by a fiber tract–based AI robot to the trial comparator while all participants receive guideline-based medical care. Eligible adults with hypertensive basal ganglia hemorrhage of 15–30 mL and measurable neurologic deficits will be enrolled and randomized across participating hospitals. The AI system uses fiber-tract information to plan and guide a robotic trajectory intended to reduce damage to critical white-matter pathways during evacuation. Primary outcomes include perioperative metrics and longer-term functional recovery measured over follow-up.

Who should consider this trial

Good fit: Adults (≥18 years) with hypertensive basal ganglia hemorrhage of 15–30 mL, GCS ≥9, pre-ICH mRS ≤1, measurable functional impairment, a stable CT scan, and for whom surgery can be performed within 72 hours are the ideal candidates.

Not a fit: Patients with hematomas involving the thalamus or brainstem, radiologic vascular abnormalities, signs of impending herniation, irreversible coagulopathy, recent anticoagulant use, pregnancy, or other exclusionary conditions are unlikely to benefit from enrollment.

Why it matters

Potential benefit: If successful, the approach could improve surgical accuracy and lead to better functional recovery and fewer complications after moderate basal ganglia hemorrhage.

How similar studies have performed: Image-guided and minimally invasive hematoma evacuation techniques have shown promising signals in prior work, but fiber tract–based AI robotic guidance is relatively novel and lacks robust randomized evidence for this specific patient group.

Eligibility criteria

Inclusion Criteria:

1. Age ≥18 years at randomization
2. Hypertensive basal ganglia hemorrhage confirmed by CT/CTA
3. Hematoma volume 15-30 mL (excluding 30 mL)
4. Functional impairment (aphasia, hemiparesis with strength ≤3, or NIHSS ≥15)
5. CT shows stable hematoma (≥6h interval, change \<5 mL)
6. GCS ≥9
7. Surgery feasible within 72h after onset
8. Pre-ICH mRS ≤1
9. Informed consent obtained according to laws and ethics

Exclusion Criteria:

1. Hematoma involving the thalamus, midbrain, or other areas
2. Radiologically diagnosed cerebrovascular abnormalities, as well as ischemic infarction converting to intracerebral hemorrhage, or recent (within 1 year) recurrence of intracerebral hemorrhage
3. Signs of impending brain herniation, such as midline shift \>1 cm or changes in the ipsilateral pupil;
4. Any irreversible coagulopathy disorders or known coagulation system disease; platelet count \<100,000; INR \>1.4; use of anticoagulants within 7 days prior to this hemorrhage;
5. Pregnant or possibly pregnant
6. Patients with severe concomitant diseases that may interfere with outcome assessment
7. Poor compliance or follow-up difficulties

Where this trial is running

Hangzhou, Zhejiang

• the Second Affiliated Hospital of Zhejiang University School of Medicine — Hangzhou, Zhejiang, China (Recruiting)

Study contacts

• Principal investigator: Gao Chen — Second Affiliated Hospital, School of Medicine, Zhejiang University
• Study coordinator: Gao Chen

Conditions Basal Ganglia Hemorrhage