JMIR Publications

ABSTRACT

Background:

The pathological and physiological state of patients with intracerebral hemorrhage (ICH) after minimally invasive surgery (MIS) is a dynamic evolution, and the traditional models cannot dynamically predict prognosis. Clinical data at multiple time points often show the characteristics of different categories, different numbers, and missing data. The existing models lack methods to deal with imbalanced data.

Objective:

This study aims to develop and validate a dynamic prognostic model using multi-timepoint data from ICH patients undergoing MIS to predict survival and functional outcomes.

Methods:

In this study, 287 patients who underwent MIS for ICH were retrospectively collected on the day of surgery, days 1, 3, 7, and 14 after surgery, and the day of drainage tube removal. Their general information, vital signs, laboratory tests, neurological function scores, head hematoma volume, and MIS-related indicators were collected. In addition, this study proposes a multi-step attention model, namely MultiStep Transformer. The model can simultaneously output three types of prediction probabilities for 30-day survival probability, 180-day survival probability, and 180-day good outcome (modified Rankin Scale, mRS 0-3) probability. Five-fold cross validation was employed to evaluate the performance of the model and compare it with mainstream models and traditional scores. The main evaluation indexes included accuracy, precision, recall, and F1 score. Attributable value analysis was conducted to assess the key predictive features.

Results:

Among 287 patients, the 30-day survival rate was 92.3%, the 180-day survival rate was 88.8%, and the 180-day favorable outcome rate was 80.3%. The MultiStep Transformer model showed a remarkable superiority over other deep learning models in predicting survival and functional outcomes. Compared to traditional scoring systems, such as the Glasgow coma scale (GCS), National Institutes of Health Stroke Scale (NIHSS), intracerebral hemorrhage (ICH) score, and functional outcome in patients with primary intracerebral hemorrhage (FUNC) score, this model attained the optimal performance.

Conclusions:

The MultiStep Transformer model proposed in this study can effectively use imbalanced data to construct a model. It possesses good dynamic prediction ability for short-term and long-term survival and functional outcome of patients with ICH undergoing MIS, providing a novel tool for individualized assessment of prognosis among patients with ICH undergoing MIS.