Deep learning enables the automation of grading histological tissue engineered cartilage images for quality control standardization

Engineered human osteoarthritic cartilage organoids

The availability of human cell-based models capturing molecular processes of cartilage degeneration can facilitate development of disease-modifying therapies for osteoarthritis [1], a currently unmet clinical need. Here, by imposing specific inflammatory challenges upon mesenchymal stromal cells at a defined stage of chondrogenesis, we engineered a human organotypic model which recapitulates main OA pathological traits such as chondrocyte hypertrophy, cartilage matrix mineralization, enhanced catabolism and mechanical stiffening. To exemplify the utility of the model, we exposed the engineered OA cartilage organoids to factors known to attenuate pathological features, including IL-1Ra, and carried out mass spectrometry-based proteomics. We identified that IL-1Ra strongly reduced production of the transcription factor CCAAT/enhancer-binding protein beta [2] and demonstrated that inhibition of the C/EBPβ-activating kinases could revert the degradative processes. Human OA cartilage organoids thus represent a relevant tool towards the discovery of new molecular drivers of cartilage degeneration and the assessment of therapeutics targeting associated pathways.

Application of Artificial Intelligence in Tissue Engineering

Tissue engineering, a crucial approach in medical research and clinical applications, aims to regenerate damaged organs. By combining stem cells, biochemical factors, and biomaterials, it encounters challenges in designing complex 3D structures. Artificial intelligence (AI) enhances tissue engineering through computational modeling, biomaterial design, cell culture optimization, and personalized medicine. This review explores AI applications in organ tissue engineering (bone, heart, nerve, skin, cartilage), employing various machine learning (ML) algorithms for data analysis, prediction, and optimization. Each section discusses common ML algorithms and specific applications, emphasizing the potential and challenges in advancing regenerative therapies.

Harnessing human adipose-derived stromal cell chondrogenesis in vitro for enhanced endochondral ossification

Endochondral ossification (ECO), the major ossification process during embryogenesis and bone repair, involves the formation of a cartilaginous template remodelled into a functional bone organ. Adipose-derived stromal cells (ASC), non-skeletal multipotent progenitors from the stromal vascular fraction (SVF) of human adipose tissue, were shown to recapitulate ECO and generate bone organs in vivo when primed into a hypertrophic cartilage tissue (HCT) in vitro. However, the reproducibility of ECO was limited and the major triggers remain unknown. We studied the effect of the expansion of cells and maturation of HCT on the induction of the ECO process. SVF cells or expanded ASC were seeded onto collagen sponges, cultured in chondrogenic medium for 3-6 weeks and implanted ectopically in nude mice to evaluate their bone-forming capacities. SVF cells from all tested donors formed mature HCT in 3 weeks whereas ASC needed 4-5 weeks. A longer induction increased the degree of maturation of the HCT, with a gradually denser cartilaginous matrix and increased mineralization. This degree of maturation was highly predictive of their bone-forming capacity in vivo, with ECO achieved only for an intermediate maturation degree. In parallel, expanding ASC also resulted in an enrichment of the stromal fraction characterized by a rapid change of their proteomic profile from a quiescent to a proliferative state. Inducing quiescence rescued their chondrogenic potential. Our findings emphasize the role of monolayer expansion and chondrogenic maturation degree of ASC on ECO and provides a simple, yet reproducible and effective approach for bone formation to be tested in specific clinical models.

Deep Learning augmented osteoarthritis grading standardization

Manual grading of cartilage histology images for investigating the extent and severity of osteoarthritis (OA) involves critical examination of the cell characteristics, which makes this task tiresome, tedious and error-prone. This results in wide inter-observer variation, causing ambiguities in OA grade prediction. Such drawbacks of manual assessment can be overcome by implementing artificial intelligence (AI) based automated image classification techniques such as deep learning (DL). Hence in this manuscript, we present the feasibility of training a deep neural network with cartilage histology images, which can grade the extent and severity of knee OA based on modified Mankin scoring system. The grading system used here for automating OA grading was simplified and modified based on the microscopic observations from the histology images, where three parameters (Safranin-O staining intensity, chondrocyte distribution & arrangement, and morphology) were considered for evaluating the OA progression. The histology images were tiled, labelled, and grouped together based on the developed grading system (Grade 0-3). Four different DL architectures were tried for image classification and the best performing model was selected by five-fold validation method. With a validation accuracy of about 84%, 0.632 Cohen's kappa score, and an excellent ROC-AUC ranging between 0.89-0.99, DenseNet121 was selected among the four models as the best performing model, and was used for inferencing on new data. Final grades obtained from the models were in accordance with the grades provided by the medical experts. We hereby demonstrate that a DL architecture can be taught to interpret the degree of cartilage degradation, with excellent discriminatory ability across all four classes of OA severity. Unlike other works where radiographic images have been considered for grading of OA, we have considered histology images which is a fundamental approach for grading OA extent and severity. This would bring a paradigm shift in histology-based assessment of OA, making this automated approach to be explored as an option for OA grading standardization.

Automated Prediction of Osteoarthritis Level in Human Osteochondral Tissue Using Histopathological Images

Osteoarthritis (OA) is the most common arthritis and the leading cause of lower extremity disability in older adults. Understanding OA progression is important in the development of patient-specific therapeutic techniques at the early stage of OA rather than at the end stage. Histopathology scoring systems are usually used to evaluate OA progress and the mechanisms involved in the development of OA. This study aims to classify the histopathological images of cartilage specimens automatically, using artificial intelligence algorithms. Hematoxylin and eosin (HE)- and safranin O and fast green (SafO)-stained images of human cartilage specimens were divided into early, mild, moderate, and severe OA. Five pre-trained convolutional networks (DarkNet-19, MobileNet, ResNet-101, NasNet) were utilized to extract the twenty features from the last fully connected layers for both scenarios of SafO and HE. Principal component analysis (PCA) and ant lion optimization (ALO) were utilized to obtain the best-weighted features. The support vector machine classifier was trained and tested based on the selected descriptors to achieve the highest accuracies of 98.04% and 97.03% in HE and SafO, respectively. Using the ALO algorithm, the F1 scores were 0.97, 0.991, 1, and 1 for the HE images and 1, 0.991, 0.97, and 1 for the SafO images for the early, mild, moderate, and severe classes, respectively. This algorithm may be a useful tool for researchers to evaluate the histopathological images of OA without the need for experts in histopathology scoring systems or the need to train new experts. Incorporating automated deep features could help to improve the characterization and understanding of OA progression and development.

Objective evaluation of chondrocyte density & cloning after joint injury using convolutional neural networks

Variations in chondrocyte density and organization in cartilage histology sections are associated with osteoarthritis progression. Rapid, accurate quantification of these two features can facilitate the evaluation of cartilage health and advance the understanding of their significance. The goal of this work was to adapt deep-learning-based methods to detect articular chondrocytes and chondrocyte clones from safranin-O-stained cartilage to evaluate chondrocyte cellularity and organization. The U-net and "you-only-look-once" (YOLO) models were trained and validated for identifying chondrocytes and chondrocyte clones, respectively. Validated models were then used to quantify chondrocyte and clone density in talar cartilage from Yucatan minipigs sacrificed 1 week, 3, 6, and 12 months after fixation of an intra-articular fracture of the hock joint. There was excellent/good agreement between expert researchers and the developed models in identifying chondrocytes/clones (U-net: R²  = 0.93, y = 0.90x-0.69; median F1 score: 0.87/YOLO: R²  = 0.79, y = 0.95x; median F1 score: 0.67). Average chondrocyte density increased 1 week after fracture (from 774 to 856 cells/mm² ), decreased substantially 3 months after fracture (610 cells/mm² ), and slowly increased 6 and 12 months after fracture (638 and 683 cells/mm² , respectively). Average detected clone density 3, 6, and 12 months after fracture (11, 11, 9 clones/mm² ) was higher than the 4-5 clones/mm² detected in normal tissue or 1 week after fracture and show local increases in clone density that varied across the joint surface with time. The accurate evaluation of cartilage cellularity and organization provided by this deep learning approach will increase objectivity of cartilage injury and regeneration assessments.

Automatic Detection of Medial and Lateral Compartments from Histological Sections of Mouse Knee Joints Using the Single-Shot Multibox Detector Algorithm

OBJECTIVE

Although mouse osteoarthritis (OA) models are widely used, their histological analysis may be susceptible to arbitrariness and inter-examiner variability in conventional methods. Therefore, a method for the unbiased scoring of OA histology is needed. In this study, as the first step for establishing this system, we developed a computer-vision algorithm that automatically detects the medial and lateral compartments of mouse knee sections in a rigorous and unbiased manner.

DESIGN

A total of 706 images of coronal sections of mouse knee joints stained by hematoxylin and eosin, safranin O, or toluidine blue were randomly divided into training and validation images at a ratio of 80:20. A model to detect both compartments automatically was built by machine learning using a single-shot multibox detector (SSD) algorithm with training images. The model was tested to determine whether it could accurately detect both compartments by analyzing the validation images and 52 images of sections stained with Picrosirius red, a method not used for the training images.

RESULTS

The trained model accurately detected both medial and lateral compartments of all 140 validation images regardless of the staining method employed, severity of articular cartilage defects, and the anatomical positions and conditions of the sections. Our model also correctly detected both compartments of 50 of 52 Picrosirius red-stained images.

CONCLUSIONS

By applying deep learning based on the SSD algorithm, we successfully developed a model that detects the locations of the medial and lateral compartments of tissue sections of mouse knee joints with high accuracy.

From Single Batch to Mass Production-Automated Platform Design Concept for a Phase II Clinical Trial Tissue Engineered Cartilage Product

Advanced Therapy Medicinal Products (ATMP) provide promising treatment options particularly for unmet clinical needs, such as progressive and chronic diseases where currently no satisfying treatment exists. Especially from the ATMP subclass of Tissue Engineered Products (TEPs), only a few have yet been translated from an academic setting to clinic and beyond. A reason for low numbers of TEPs in current clinical trials and one main key hurdle for TEPs is the cost and labor-intensive manufacturing process. Manual production steps require experienced personnel, are challenging to standardize and to scale up. Automated manufacturing has the potential to overcome these challenges, toward an increasing cost-effectiveness. One major obstacle for automation is the control and risk prevention of cross contaminations, especially when handling parallel production lines of different patient material. These critical steps necessitate validated effective and efficient cleaning procedures in an automated system. In this perspective, possible technologies, concepts and solutions to existing ATMP manufacturing hurdles are discussed on the example of a late clinical phase II trial TEP. In compliance to Good Manufacturing Practice (GMP) guidelines, we propose a dual arm robot based isolator approach. Our novel concept enables complete process automation for adherent cell culture, and the translation of all manual process steps with standard laboratory equipment. Moreover, we discuss novel solutions for automated cleaning, without the need for human intervention. Consequently, our automation concept offers the unique chance to scale up production while becoming more cost-effective, which will ultimately increase TEP availability to a broader number of patients.