Automated detection and growth tracking of 3D bio-printed organoid clusters using optical coherence tomography with deep convolutional neural networks

Abstract

Organoids are advancing the development of accurate prediction of drug efficacy and toxicity in vitro. These advancements are attributed to the ability of organoids to recapitulate key structural and functional features of organs and parent tumor. Specifically, organoids are self-organized assembly with a multi-scale structure of 30–800 μm, which exacerbates the difficulty of non-destructive three-dimensional (3D) imaging, tracking and classification analysis for organoid clusters by traditional microscopy techniques. Here, we devise a 3D imaging, segmentation and analysis method based on Optical coherence tomography (OCT) technology and deep convolutional neural networks (CNNs) for printed organoid clusters (Organoid Printing and optical coherence tomography-based analysis, OPO). The results demonstrate that the organoid scale influences the segmentation effect of the neural network. The multi-scale information-guided optimized EGO-Net we designed achieves the best results, especially showing better recognition workout for the biologically significant organoid with diameter ≥50 μm than other neural networks. Moreover, OPO achieves to reconstruct the multiscale structure of organoid clusters within printed microbeads and calibrate the printing errors by segmenting the printed microbeads edges. Overall, the classification, tracking and quantitative analysis based on image reveal that the growth process of organoid undergoes morphological changes such as volume growth, cavity creation and fusion, and quantitative calculation of the volume demonstrates that the growth rate of organoid is associated with the initial scale. The new method we proposed enable the study of growth, structural evolution and heterogeneity for the organoid cluster, which is valuable for drug screening and tumor drug sensitivity detection based on organoids.

Segmentation and Multi-Timepoint Tracking of 3D Cancer Organoids from Optical Coherence Tomography Images Using Deep Neural Networks

“…On the other hand, the test set included multiple days that were not present in the training and validation set, specifically in the early days right after the initial cultivation. This provides an intrinsic hurdle to overcome for the segmentation network, yet the performance results for the early days are in line with previous studies that have shown that smaller organoids are more challenging to segment accurately due to their small size and similarities to background noise [21]. The segmentation performances obtained on the days that were also included in the samples for the training and validation sets were, as expected, satisfactory (DICE = 0.80).…”

“…The accuracy of organoid segmentation is influenced by factors such as size, shape, and internal structure [ 21 ]. Hence, it is important to mention that the training and validation of the segmentation network were conducted on samples acquired from day 5 to day 13 after the initial organoid cultivation.…”

“…Then, the spherical volume is compared to a reference volume of a sphere with a diameter of 25 μm. Any objects having a volume smaller than this reference spherical volume are removed [ 21 ].…”

“…In the context of organoid segmentation in OCT images, an initial study proposed by Gil et al in 2021 [ 8 ] demonstrated the feasibility of volumetrically tracking organoid growth in patient-derived cancer organoids. The authors proposed an automated image analysis framework based on preprocessing, registration, segmentation using k-means clustering, and a centroid particle tracking algorithm to track individual organoids over multiple timepoints covering 48 h. Recently, Bao et al [ 21 ] employed deep convolutional neural networks for organoid segmentation and demonstrated the feasibility of detecting and tracking the growth of 3D bio-printed organoid clusters in OCT volumes using a deep learning approach. The authors imaged the organoids over 7 days at an interval of every 24 h, and the results of the study also demonstrated the influence of the organoid scale on segmentation accuracy.…”

Enhancing quality control in bioprinting through machine learning

“…Finally, as can be seen in table 2, ML has been studied to automatically extract relevant features related to the state of bioprinted cells from image data [104,109]. In an interesting application, Yao et al proposed an unsupervised algorithm based on generative adversarial networks (GANs) for the automatic segmentation of cell nuclei in 3D scaffolds (figure 6(a)).…”