Live Neuronal Coculture Metabolic Profiles via Deep Learning-Augmented Raman Imaging

A dataset from figshare published by Li-En Lin on May 28, 2026, featuring metabolic imaging data from live neuronal cocultures. The data was generated using a tandem deep-learning approach with stimulated Raman scattering microscopy to profile lipid, protein, glucose, and D2O metabolism. It captures cell-type-specific metabolic dynamics in neurons, astrocytes, and oligodendrocytes under physiological and pathological conditions.

Use Cases

• Train cell segmentation models based on the U-Net architecture mentioned in the description.
• Analyze metabolic flux in different neuronal cell types based on deuterium-labeled metabolite data.
• Model metabolic adaptations to pathological conditions like Huntington's disease or proteasome inhibition described in the study.
• Benchmark high-resolution 3D imaging reconstruction algorithms based on the RCNN framework used.

Strengths

• Data is generated using a noninvasive, high-speed stimulated Raman scattering microscopy method designed for live cells.
• Profiles four specific metabolic pathways: lipid, protein, glucose, and D2O metabolism.
• Includes observations under multiple defined conditions: NMDA receptor activation, proteasome inhibition, and a Huntington's disease model.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is small at 5.4 MB, which may limit the scope of analysis.

Provenance

Source

figshare

Collection Method

Generated via deep-learning-augmented stimulated Raman scattering microscopy on live neuronal cocultures.

Freshness

Last updated 2026-05-28 19:44:38; freshness should be verified.