Work

Liver MRI Methods

I built this small introductory project to understand methods developed for patient-level clinical data in the world of multiphasic MRI.

Training and validation binary cross-entropy fall across the three U-Net epochs.
Training and validation binary cross-entropy plotted by epochSourceGross et al., LiverHccSeg v1.1TCGA-LIHC, The Cancer Imaging ArchiveCC BY 4.0

What I was asked

While patient separation matters, image geometry, registration and annotation choices shape the measurement before a model sees it.

What I did

  1. Geometry

    Audit phase availability, orientation, spacing and physical grids before comparing voxels across scans.

  2. Registration

    I compared the dataset-supplied registration with a seeded rigid SimpleITK baseline on three examinations.

  3. Logistic baseline

    Separate patients before fitting a logistic voxel classifier using four declared dynamic-contrast signal features.

  4. U-Net

    Run a fixed three-epoch, four-channel 2D U-Net smoke test and review its predictions alongside its numerical scores.

  5. Results

    Preserve the failed baseline, configuration and result tables so later debugging studies extend the record instead of rewriting it.

What it changed

The logistic baseline could rank tumour regions strongly while labelling too much liver.

The U-Net produced apparently stronger overlap for one held-out case, but the evaluation units differed and the images told the more important story: broad liver-shaped predictions and false positives in a declared negative examination.

Before making the network larger, the project now needs to test target construction, slice balance and the ways the manual liver mask may have created a shortcut.