Cancer characterization using light backscattering spectroscopy and quantitative ultrasound - an ex vivo study on sarcoma subtypes

Sep 24, 2026

Uploaded by McQueen

Histological analysis is a gold standard for cancer diagnosis, but it can be affected by subjectivity and sampling bias. A quantitative bimodal approach combines light backscattering spectroscopy with quantitative ultrasound to guide assessment of suspicious regions in two bone tumor types from animal models: chondrosarcomas and osteosarcomas. Ultrasound and optical parameters show significant differences, enabling estimation of morphological features at cellular and nuclear scales with reported relative errors, and providing complementary tissue-property information for tumor characterization.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Introduction

Histological diagnosis and limitations

Clinical need for non-invasive quantitative guidance

Rationale for bimodal light-ultrasound approach

Methods and imaging principles

Quantitative ultrasound techniques (BSC, envelope statistics)

Optical techniques (EBS, LSS) and expected complementarity

Spectral and envelope-based analysis of RF signals

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN Cancer characterization using light backscattering spectroscopy and quantitative ultrasound: an ex vivo study on sarcoma subtypes Cyril Malinet1*, Bruno Montcel1, Aurélie Dutour2, Iveta Fajnorova2, Hervé Liebgott1 & Pauline Muleki‑Seya1 Histological analysis is the gold standard method for cancer diagnosis. However, it is prone to subjectivity and sampling bias. In response to these limitations, we introduce a quantitative bimodal approach that aims to provide non‑invasive guidance towards suspicious regions. Light backscattering spectroscopy and quantitative ultrasound techniques were combined to characterize two different bone tumor types from animal models: chondrosarcomas and osteosarcomas. Two different cell lines were used to induce osteosarcoma growth. Histological analyses were conducted to serve as references. Three ultrasound parameters and intensities of the light reflectance profiles showed significant differences between chondrosarcomas and osteosarcomas atthe 5% level. Likewise, variations in the same biomarkers were reported for the two types ofosteosarcoma, despite their similar morphology observed in the histological examinations. These observations show the sensitivity of our techniques in probing fine tissue properties. Secondly, the ultrasound spectral‑ based technique identified the mean size of chondrosarcoma cells and nuclei with relative errors of about 22% and 9% respectively. The optical equivalent technique correctly extracted scatterer size distributions that encompass nuclei and cells for chondrosarcomas and osteosarcomas ( 􀀞􀀝􀀜􀀞 and 􀀞􀀝􀀜􀀛 respectively). The optical scattering contributions of nuclei were estimated at 52% forthe chondrosarcomas and 69% forthe osteosarcomas, probably indicating the abundant and the absent extracellular matrix respectively. Thus, the ultrasound and the optical methods brought complementary parameters. They successfully estimated morphological parameters atthe cellular and the nuclear scales, making our bimodal technique promising for tumor characterization. In clinical settings, the process of cancer characterization involves the determination of the tumor histological subtype and the grade, which reflects the cancer aggressiveness1. Determining these characteristics plays a crucial role in establishing the patient’s diagnosis and planning the right treatment options. This histological classification relies on cellular morphological measurements and is traditionally determined through microscopic examination, referred to as histological analyses. Although histo-cytopathology serves as a gold standard for diagnosing cancers, this technique is inherently invasive and resource-intensive. Additionally, it is subject to inter-observer and intra-observer variabilities2. For instance, the sampled sections may not include the most aggressive cancerous regions due to tumor heterogeneities. Thus, the sampling bias could lead to inaccurate diagnostics that ultimately worsen the patient’s prognosis. Other applications involved in the cancer patient care workflow suffer from subjectivity, such as intra-operative margin assessment3. Indeed, insufficient margins resulting in the presence of remaining cancerous cells after resection surgery can lead to cancer recurrence. Consequently, guiding the clinicians toward the most suspicious regions using a non-invasive quantitative tool would be of great benefit for different clinical procedures and could subsequently improve the patient outcomes. 1Université de Lyon, CREATIS, CNRS UMR 5220, Inserm U1044, INSA-Lyon, Université Lyon 1, Lyon, France. 2Centre de Recherche en Cancérologie de Lyon/Centre Léon Bérard, Equipe mort cellulaire et cancerspédiatriques, UMR INSERM 1052, CNRS 5286, Lyon, France.* email: [cyril.malinet@outlook.fr](cyril.malinet@outlook.fr) [www. nature.com/scientificreports/](www. nature.com/scientificreports/) As a starting poin

FAQs

Why is histological analysis considered the gold standard, and what limitations does it have?

Histological analysis is widely used for diagnosing cancers, but it is prone to subjectivity and sampling bias. Tumor heterogeneity and observer variability can also lead to inaccurate diagnostics.

What does the study combine to characterize sarcoma subtypes?

The study combines light backscattering spectroscopy with quantitative ultrasound. The goal is to extract complementary information about microstructures from each modality.

How do the authors assess differences between chondrosarcomas and osteosarcomas?

They report differences using ultrasound parameters and light reflectance/spectral-derived biomarkers. They also estimate cell and nuclear-related size and scattering contributions to distinguish the tumor types.

Sep 24, 2026

Uploaded by McQueen