# Age Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks

## TL;DR

This paper shows that using CNNs with hyperspectral images can more accurately predict hematoma age than traditional methods, while also reducing the data needed.

## Contribution

The study introduces a CNN that combines spectral and spatial data and identifies a minimal set of informative wavelengths for accurate hematoma age estimation.

## Key findings

- The CNN reduced the mean absolute error in hematoma age estimation from 3.24 to 2.29 days.
- A subset of 20 wavelengths achieved accuracy comparable to the full 204-band model across all hematoma stages.
- Spectral-spatial modeling and band selection improve accuracy while reducing data dimensionality.

## Abstract

Accurate estimation of hematoma age remains a major challenge in forensic practice, as current assessments rely heavily on subjective visual interpretation. Hyperspectral imaging (HSI) captures rich spectral signatures that may reflect the biochemical evolution of hematomas over time. This study evaluates whether a convolutional neural network (CNN) integrating both spectral and spatial information improves hematoma age estimation accuracy. Additionally, we investigate whether performance can be maintained using a reduced, physiologically motivated subset of wavelengths. Using a dataset of forearm hematomas from 25 participants, we applied radiometric normalization and SAM-based segmentation to extract 64×64×204 hyperspectral patches. In leave-one-subject-out cross-validation, the CNN outperformed a spectral-only Lasso baseline, reducing the mean absolute error (MAE) from 3.24 days to 2.29 days. Band-importance analysis combining SmoothGrad and occlusion sensitivity identified 20 highly informative wavelengths; using only these bands matched or exceeded the accuracy of the full 204-band model across early, middle, and late hematoma stages. These results demonstrate that spectral–spatial modeling and physiologically grounded band selection can enhance estimation accuracy while significantly reducing data dimensionality. This approach supports the development of compact multispectral systems for objective clinical and forensic evaluation.