Related Experiment Video
Updated: Aug 6, 2026
Longitudinal Morphological and Physiological Monitoring of Three-dimensional Tumor Spheroids Using Optical Coherence Tomography
Published on: February 9, 2019
A 3-dimensional Resnet model for assessment of drug efficacy in 3D cancer models using optical coherence tomography
Gavrielle R Untracht1, Jan Kaminski2, Eike Guldenring2
1Department of Health Technology, Technical University of Denmark, Kongens Lyngby, Denmark.
Abstract:
Ninety percent of drugs fail during clinical trials, mainly due to lack of clinical efficacy. Recent developments in in vitro models such as 3D tumor heterospheroids have led to improvements in failure rates, but the relative lack of standardized evaluation methods for 3D cultures limits their utility in high-throughput screening. Optical coherence tomography (OCT) shows significant promise for high-throughput screening of 3D models; however, the optimal classification model and key image features for assessing drug efficacy in OCT images of spheroids has yet to be explored in detail. In this study, we investigate whether OCT combined with machine learning methods can be used to identify biomarkers of drug efficacy in 3D tumor spheroid models. We further compare the performance of two different models to determine the optimal configuration for accurate classification. Volumetric OCT images were acquired of co-cultured HT29 spheroids treated with 3 different concentrations of cisplatin. A two-dimensional multi-view ResNet model and a three-dimensional ResNet model were used to classify the images and to identify key image features associated with each group. Differences between spheroids treated with different concentrations of cisplatin are clearly visible in the OCT images. Our model was able to classify the images based on cisplatin concentration with 71.9% accuracy using the 2D multi-view model and 91.2% accuracy using the 3D model. Key features in the 3D model significantly improved the model accuracy. These results underscore the possibility that OCT could be used for high-throughput screening of drugs using 3D in vitro models and highlight key identifying features for further investigation.
Insights
Optical coherence tomography (OCT) combined with 3D machine learning models can accurately assess drug efficacy in 3D tumor spheroids. This approach shows promise for high-throughput drug screening using advanced in vitro models.
Area of Science:
- Biomedical Engineering
- Oncology
- Medical Imaging
Background:
- High drug failure rates in clinical trials (90%) are primarily due to lack of efficacy.
- Advanced in vitro models like 3D tumor spheroids improve drug testing, but standardized evaluation methods are lacking.
- Optical coherence tomography (OCT) offers potential for high-throughput screening of 3D models, yet optimal classification models and image features for drug efficacy assessment require further study.
Purpose of the Study:
- To investigate the use of OCT and machine learning for identifying drug efficacy biomarkers in 3D tumor spheroid models.
- To compare the performance of 2D multi-view and 3D ResNet models for classifying spheroid responses to cisplatin.
- To identify key image features that indicate drug efficacy in OCT images of 3D tumor spheroids.
Main Methods:
- Volumetric OCT imaging was performed on co-cultured HT29 spheroids treated with varying cisplatin concentrations.
- Two machine learning models, a 2D multi-view ResNet and a 3D ResNet, were employed for image classification.
- Key image features associated with different treatment groups were identified using the 3D model.
Main Results:
- Visible differences in spheroid morphology were observed in OCT images based on cisplatin concentration.
- The 3D ResNet model achieved 91.2% accuracy in classifying spheroids by cisplatin concentration, outperforming the 2D multi-view model (71.9%).
- Key features identified by the 3D model significantly enhanced classification accuracy, highlighting their importance in assessing drug response.
Conclusions:
- OCT combined with 3D machine learning offers a viable method for high-throughput drug screening using 3D in vitro tumor models.
- The study identified critical image features within OCT data that correlate with drug efficacy.
- This approach has the potential to improve drug development pipelines by enabling more accurate and efficient preclinical testing.
More Related Videos
10:13Non-Destructive Evaluation of Regional Cell Density Within Tumor Aggregates Following Drug Treatment
Published on: June 21, 2022
13:34A Combined 3D Tissue Engineered In Vitro/In Silico Lung Tumor Model for Predicting Drug Effectiveness in Specific Mutational Backgrounds
Published on: April 6, 2016