Skip to content

Development of a Multimodal Deep Learning Model for Early Detection of Infectious Diseases Based on Complete Blood Count Data

Development of a Multimodal Deep Learning Model for Early Detection of Infectious Diseases Based on Complete Blood Count Data

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600122102
Enrollment
Unknown
Registered
2026-04-09
Start date
2026-04-30
Completion date
Unknown
Last updated
2026-04-14

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

Conditions

Infectious Diseases

Interventions

Gold Standard:Etiological evidence (e.g., microbiological culture, polymerase chain reaction PCR)
Index test:This is a diagnostic model constructed based on machine learning algorithms that integrates multiple parameters of the complete blood count, derived indicators, and patient baseline informa

Sponsors

West China Hospital, Sichuan University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: 1.Age: Adult patients aged >=18 years. 2.Clinical Symptoms: Patients presenting for the first time at the emergency department or fever clinic with acute fever (body temperature >=38.0°C) and/or suspected infectious symptoms (e.g., chills, cough, expectoration, urinary tract irritation symptoms, etc.). 3.Testing Requirement: Completion of a complete blood count (CBC) test within 24 hours of the visit, with the test results being complete and available for use. 4.Diagnostic Gold Standard: Possession of a definitive final clinical diagnosis that can serve as the "gold standard" for model training. The diagnosis must be based on a combination of etiological evidence (e.g., culture, PCR), serological tests, imaging examinations, and clinical response to treatment, and must be jointly confirmed by at least two senior physicians. The diagnostic outcome must clearly categorize the patient into one of the following three groups: bacterial infection group, viral infection group, or non-infectious disease group.

Exclusion criteria

Exclusion criteria: 1.Missing Information: Patients with missing or incompletely recorded key clinical information (e.g., vital signs, final diagnosis). 2.Recent Specific Treatment History: Patients who have undergone blood transfusion, chemotherapy, or radiotherapy within 14 days prior to enrollment. 3.Specific Disease History: Patients with known hematological diseases (e.g., leukemia, aplastic anemia) or hypersplenism/having undergone splenectomy.

Design outcomes

Primary

MeasureTime frame
Complete Blood Count Five-Part Differential Scatter Plot (DIFF);

Secondary

MeasureTime frame
Complete Blood Count Parameters (WBC);

Countries

China

Contacts

Public ContactFu Yang

West China Hospital, Sichuan University

fuyang827@wchscu.cn+86 189 8060 0695

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Apr 17, 2026