Inverse Protocol Prediction from Spheroid Microscopy Imaging via Morphology-Aware Structured Learning

We introduce Inverse Protocol Prediction (IPP), which is a task of inferring experimental culture conditions directly from a single bright-field spheroid image. We formulate IPP as a structured multi-label prediction problem and propose a protocol-aware learning framework that integrates morphology extraction, multimodal representation learning, and dependency-aware inference. Morphometric descriptors derived from automated spheroid segmentation are fused with deep visual embeddings via a morphometry vision fusion module. To capture biological and procedural dependencies, we develop a Hierarchical Multi-Task Transformer that conditions predictions across protocol attributes. The framework is trained with domainadversarial supervision and morphology-preserving augmentation to improve robustness to acquisition variability. Hybrid convolution attention encoders achieve the best performance, reaching 95.7% multi-attribute accuracy. We further evaluate cross dataset transfer and temporal morphology forecasting. Results demonstrate that structured, dependency-aware modeling enables reliable reconstruction of experimental protocols from imaging alone, supporting reproducibility auditing and protocol validation in 3D cell culture systems.