Deep learning applied to the histopathological diagnosis of ameloblastomas and ameloblastic carcinomas

BACKGROUND

Odontogenic tumors (OT) are composed of heterogeneous lesions, which can be benign or malignant, with different behavior and histology. Within this classification, ameloblastoma and ameloblastic carcinoma (AC) represent a diagnostic challenge in daily histopathological practice due to their similar characteristics and the limitations that incisional biopsies represent. From these premises, we wanted to test the usefulness of models based on artificial intelligence (AI) in the field of oral and maxillofacial pathology for differential diagnosis. The main advantages of integrating Machine Learning (ML) with microscopic and radiographic imaging is the ability to significantly reduce intra-and inter observer variability and improve diagnostic objectivity and reproducibility.

METHODS

Thirty Digitized slides were collected from different diagnostic centers of oral pathology in Brazil. After performing manual annotation in the region of interest, the images were segmented and fragmented into small patches. In the supervised learning methodology for image classification, three models (ResNet50, DenseNet, and VGG16) were focus of investigation to provide the probability of an image being classified as class0 (i.e., ameloblastoma) or class1 (i.e., Ameloblastic carcinoma).

RESULTS

The training and validation metrics did not show convergence, characterizing overfitting. However, the test results were satisfactory, with an average for ResNet50 of 0.75, 0.71, 0.84, 0.65, and 0.77 for accuracy, precision, sensitivity, specificity, and F1-score, respectively.

CONCLUSIONS

The models demonstrated a strong potential of learning, but lack of generalization ability. The models learn fast, reaching a training accuracy of 98%. The evaluation process showed instability in validation; however, acceptable performance in the testing process, which may be due to the small data set. This first investigation opens an opportunity for expanding collaboration to incorporate more complementary data; as well as, developing and evaluating new alternative models.

Low-Grade Myofibroblastic Sarcoma of the Oral and Maxillofacial Region: An International Clinicopathologic Study of 13 Cases and Literature Review

Low-grade myofibroblastic sarcoma (LGMS) represents an atypical tumor composed of myofibroblasts with a variety of histological patterns and with a high tendency to local recurrence and a low probability of distant metastases. LGMS has predilection for the head and neck regions, especially the oral cavity. This study aimed to report 13 new cases of LGMS arising in the oral and maxillofacial region. This study included LGMS cases from five oral and maxillofacial pathology laboratories in four different countries (Brazil, Peru, Guatemala, and South Africa). Their clinical, radiographic, histopathological, and immunohistochemical findings were evaluated. In this current international case series, most patients were females with a mean age of 38.7 years, and commonly presenting a nodular lesion in maxilla. Microscopically, all cases showed a neoplasm formed by oval to spindle cells in a fibrous stroma with myxoid and dense areas, some atypical mitoses, and prominent nucleoli. The immunohistochemical panel showed positivity for smooth muscle actin (12 of 13 cases), HHF35 (2 of 4 cases), β-catenin (3 of 5 cases), desmin (3 of 11 cases), and Ki-67 (range from 5 to 50%). H-caldesmon was negative for all cases. The diagnosis of LGMS was confirmed in all cases. LGMS shows predominance in young adults, with a slight predilection for the female sex, and maxillary region. LGMS should be a differential diagnosis of myofibroblastic lesions that show a proliferation of spindle cells in a fibrous stroma with myxoid and dense areas and some atypical mitoses, supporting the diagnosis with a complementary immunohistochemical study. Complete surgical excision with clear margins is the treatment of choice. However, long-term follow-up information is required before definitive conclusions can be drawn regarding the incidence of recurrence and the possibility of metastasis.

Development and Evaluation of a Convolutional Neural Network for Microscopic Diagnosis Between Pleomorphic Adenoma and Carcinoma Ex-Pleomorphic Adenoma

AIMS

To develop a model capable of distinguishing carcinoma ex-pleomorphic adenoma from pleomorphic adenoma using a convolutional neural network architecture.

METHODS AND RESULTS

A cohort of 83 Brazilian patients, divided into carcinoma ex-pleomorphic adenoma (n = 42) and pleomorphic adenoma (n = 41), was used for training a convolutional neural network. The whole-slide images were annotated and fragmented into 743 869 (carcinoma ex-pleomorphic adenomas) and 211 714 (pleomorphic adenomas) patches, measuring 224 × 224 pixels. Training (80%), validation (10%), and test (10%) subsets were established. The Residual Neural Network (ResNet)-50 was chosen for its recognition and classification capabilities. The training and validation graphs, and parameters derived from the confusion matrix, were evaluated. The loss curve recorded 0.63, and the accuracy reached 0.93. Evaluated parameters included specificity (0.88), sensitivity (0.94), precision (0.96), F1 score (0.95), and area under the curve (0.97).

CONCLUSIONS

The study underscores the potential of ResNet-50 in the microscopic diagnosis of carcinoma ex-pleomorphic adenoma. The developed model demonstrated strong learning potential, but exhibited partial limitations in generalization, as indicated by the validation curve. In summary, the study established a promising baseline despite limitations in model generalization. This indicates the need to refine methodologies, investigate new models, incorporate larger datasets, and encourage inter-institutional collaboration for comprehensive studies in salivary gland tumors.