RamanSTS dataset, a large collection of Raman spectra of soft tissue sarcomas

Microscopic Raman Spectroscopy of Soft Tissue Sarcomas Soft tissue sarcomas (STS) are a diverse and rare group of malignant tumors arising from the connective tissues of the body, including fibrous tissue, muscles, fat, nerves, and blood vessels. The heterogeneity and infrequency of these tumors pose significant challenges in both diagnosis and treatment. Surgical resection remains the primary treatment strategy, often complemented by radiation or chemotherapy, contingent upon the tumor's size, location, and stage. However, current methods for assessing intraoperative margins are limited, underscoring the need for improved approaches that enhance both efficiency and accuracy. This dataset comprises 286,672 ex-vivo Raman spectra acquired from surgical samples of seven patients using a 633 nm excitation laser. The tissue types include both malignant and benign tumors, as well as normal tissue. Spectra were collected from 100 µm × 100 µm regions using a 51×51 grid with 0.5-second exposure per spot. Each individual Raman spectrum spans the biochemical fingerprint region from 400 cm−1 to 1800 cm−1, consisting of 1,401 data points per spectrum. All tissue categories adhere to the World Health Organization (WHO) classification of soft tissue tumors. Sample types and abbreviations include: MSC: Muscle (normal) SKN: Skin layers (normal) FAT: Fat (normal) LEM: Leiomyoma (benign) MLS: Myxoid liposarcoma (malignant) PLS: High-grade pleomorphic liposarcoma (malignant) LEI: Leiomyosarcoma (malignant) HMS: High-grade myxofibrosarcoma (malignant) The dataset is divided into two compressed files: RamanSTS_Part1.zip – 13 .npy files (137,752 spectra) RamanSTS_Part2.zip – 5 .npy files (148,920 spectra) Each file contains Raman spectra associated with specific tissue types. Spectra can be tracked using the Idx column provided in the accompanying metadata file. For implementation details, please visit the accompanying GitHub repository. Please cite the following publication when using this dataset: Boroji M. et al., (2025) "Ex-Vivo Raman Spectroscopy and AI-Based Classification of Soft Tissue Sarcomas", PLOS ONE 20(9): e0330618.