Multicentric thrombus segmentation using an attention-based recurrent network with gradual modality dropout

Dimensions overall score 3.0

• An attention-based recurrent network with gradual modality dropout for multicentric thrombus segmentation in 3D brain scans. We introduce a methodology that couples an attention-based recurrent segmentation network (UpAttLLSTM), a training schedule that progressively increases the difficulty of hetero-modal learning, with gradual modality dropout, UpAttLLSTM aggregates context across slices via recurrent units (2.5D) and uses attention gates to fuse complementary cues across available sequences, making it robust to anisotropy and class imbalance.problem 35%
  Evidencepartial

  An attention-based recurrent network with gradual modality dropout for multicentric thrombus segmentation in 3D brain scans. We introduce a methodology that couples an attention-based recurrent segmentation network (UpAttLLSTM), a training schedule that progressively increases the difficulty of hetero-modal learning, with gradual modality dropout, UpAttLLSTM aggregates context across slices via recurrent units (2.5D) and uses attention gates to fuse complementary cues across available sequences, making it robust to anisotropy and class imbalance.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• Detecting and delineating tiny targets in 3D brain scans is a central yet under-addressed challenge in medical imaging.In ischemic stroke, for instance, the culprit thrombus is small, low-contrast, and variably expressed across modalities(e.g., susceptibility-weighted T2 blooming, diffusion restriction on DWI/ADC), while real-world multi-center dataintroduce domain shifts, anisotropy, and frequent missing sequences. We introduce a methodology that couples an attention-based recurrent segmentation network (UpAttLLSTM), a training schedule that progressively increases the difficulty of hetero-modal learning, with gradual modality dropout, UpAttLLSTM aggregates context across slices via recurrent units (2.5D) and uses attention gates to fuse complementary cues across available sequences, making it robust to anisotropy and class imbalance.method 35%
  Evidencepartial

  Detecting and delineating tiny targets in 3D brain scans is a central yet under-addressed challenge in medical imaging.In ischemic stroke, for instance, the culprit thrombus is small, low-contrast, and variably expressed across modalities(e.g., susceptibility-weighted T2 blooming, diffusion restriction on DWI/ADC), while real-world multi-center dataintroduce domain shifts, anisotropy, and frequent missing sequences. We introduce a methodology that couples an attention-based recurrent segmentation network (UpAttLLSTM), a training schedule that progressively increases the difficulty of hetero-modal learning, with gradual modality dropout, UpAttLLSTM aggregates context across slices via recurrent units (2.5D) and uses attention gates to fuse complementary cues across available sequences, making it robust to anisotropy and class imbalance.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• ScienceToStartup currently rates this 3.0/10 on the public viability pass. Gradual modality dropout systematically simulates site heterogeneity,noise, and missing modalities during training, acting as both augmentation and regularization to improve multi-center generalization.result 35%
  Evidencepartial

  ScienceToStartup currently rates this 3.0/10 on the public viability pass. Gradual modality dropout systematically simulates site heterogeneity,noise, and missing modalities during training, acting as both augmentation and regularization to improve multi-center generalization.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• Medical Imaging Segmentation moved forward this cycle; last verified April 2026. Public score 3.0/10.why now 35%
  Evidencepartial

  Medical Imaging Segmentation moved forward this cycle; last verified April 2026. Public score 3.0/10.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

Competitive landscape

An attention-based recurrent network with gradual modality dropout for multicentric thrombus segmentation in 3D brain scans.