RESUMO
BACKGROUND: Neural tumors are difficult to distinguish based solely on cellularity and often require immunohistochemical staining to aid in identifying the cell lineage. This article investigates the potential of a Convolutional Neural Network for the histopathological classification of the three most prevalent benign neural tumor types: neurofibroma, perineurioma, and schwannoma. METHODS: A model was developed, trained, and evaluated for classification using the ResNet-50 architecture, with a database of 30 whole-slide images stained in hematoxylin and eosin (106, 782 patches were generated from and divided among the training, validation, and testing subsets, with strategies to avoid data leakage). RESULTS: The model achieved an accuracy of 70% (64% normalized), and showed satisfactory results for differentiating two of the three classes, reaching approximately 97% and 77% as true positives for neurofibroma and schwannoma classes, respectively, and only 7% for perineurioma class. The AUROC curves for neurofibroma and schwannoma classes was 0.83%, and 0.74% for perineurioma. However, the specificity rate for the perineurioma class was greater (83%) than in the other two classes (neurofibroma with 61%, and schwannoma with 60%). CONCLUSION: This investigation demonstrated significant potential for proficient performance with a limitation regarding the perineurioma class (the limited feature variability observed contributed to a lower performance).
Assuntos
Estudos de Viabilidade , Neoplasias Bucais , Neoplasias de Bainha Neural , Redes Neurais de Computação , Neurilemoma , Neurofibroma , Humanos , Neurofibroma/patologia , Neurilemoma/patologia , Neoplasias de Bainha Neural/patologia , Neoplasias Bucais/patologia , Diagnóstico DiferencialRESUMO
BACKGROUND: Dysplasia grading systems for oral epithelial dysplasia are a source of disagreement among pathologists. Therefore, machine learning approaches are being developed to mitigate this issue. METHODS: This cross-sectional study included a cohort of 82 patients with oral potentially malignant disorders and correspondent 98 hematoxylin and eosin-stained whole slide images with biopsied-proven dysplasia. All whole-slide images were manually annotated based on the binary system for oral epithelial dysplasia. The annotated regions of interest were segmented and fragmented into small patches and non-randomly sampled into training/validation and test subsets. The training/validation data were color augmented, resulting in a total of 81,786 patches for training. The held-out independent test set enrolled a total of 4,486 patches. Seven state-of-the-art convolutional neural networks were trained, validated, and tested with the same dataset. RESULTS: The models presented a high learning rate, yet very low generalization potential. At the model development, VGG16 performed the best, but with massive overfitting. In the test set, VGG16 presented the best accuracy, sensitivity, specificity, and area under the curve (62%, 62%, 66%, and 65%, respectively), associated with the higher loss among all Convolutional Neural Networks (CNNs) tested. EfficientB0 has comparable metrics and the lowest loss among all convolutional neural networks, being a great candidate for further studies. CONCLUSION: The models were not able to generalize enough to be applied in real-life datasets due to an overlapping of features between the two classes (i.e., high risk and low risk of malignization).
Assuntos
Aprendizado Profundo , Humanos , Estudos Transversais , Redes Neurais de Computação , Aprendizado de Máquina , BiópsiaRESUMO
Studies suggest non-invasive transcutaneous auricular vagus nerve stimulation (taVNS) as a potential therapeutic option for various pathological conditions, such as epilepsy and depression. Exhalation-controlled taVNS, which synchronizes stimulation with internal body rhythms, holds promise for enhanced neuromodulation, but there is no closed-loop system in the literature capable of performing such integration in real time. In this context, the objective was to develop real-time signal processing techniques and an integrated closed-loop device with sensors to acquire physiological data. After a conditioning stage, the signal is processed and delivers synchronized electrical stimulation during the patient's expiratory phase. Additional modules were designed for processing, software-controlled selectors, remote and autonomous operation, improved analysis, and graphical visualization. The signal processing method effectively extracted respiratory cycles and successfully attenuated signal noise. Heart rate variability was assessed in real time, using linear statistical evaluation. The prototype feedback stimulator device was physically constructed. Respiratory peak detection achieved an accuracy of 90%, and the real-time processing resulted in a small delay of up to 150 ms in the detection of the expiratory phase. Thus, preliminary results show promising accuracy, indicating the need for additional tests to optimize real-time processing and the application of the prototype in clinical studies.
RESUMO
There is a great deal of evidence showing the capacity of physical exercise to enhance cognitive function, reduce anxiety and depression, and protect the brain against neurodegenerative disorders. Although the effects of exercise are well documented in the mature brain, the influence of exercise in the developing brain has been poorly explored. Therefore, we investigated the morphological and functional hippocampal changes in adult rats submitted to daily treadmill exercise during the adolescent period. Male Wistar rats aged 21 postnatal days old (P21) were divided into two groups: exercise and control. Animals in the exercise group were submitted to daily exercise on the treadmill between P21 and P60. Running time and speed gradually increased over this period, reaching a maximum of 18 m/min for 60 min. After the aerobic exercise program (P60), histological and behavioral (water maze) analyses were performed. The results show that early-life exercise increased mossy fibers density and hippocampal expression of brain-derived neurotrophic factor and its receptor tropomyosin-related kinase B, improved spatial learning and memory, and enhanced capacity to evoke spatial memories in later stages (when measured at P96). It is important to point out that while physical exercise induces hippocampal plasticity, degenerative effects could appear in undue conditions of physical or psychological stress. In this regard, we also showed that the exercise protocol used here did not induce inflammatory response and degenerating neurons in the hippocampal formation of developing rats. Our findings demonstrate that physical exercise during postnatal development results in positive changes for the hippocampal formation, both in structure and function.
Assuntos
Hipocampo/citologia , Hipocampo/fisiologia , Memória/fisiologia , Plasticidade Neuronal/fisiologia , Condicionamento Físico Animal/fisiologia , Animais , Western Blotting , Contagem de Células , Ensaio de Imunoadsorção Enzimática , Imunofluorescência , Imuno-Histoquímica , Masculino , Aprendizagem em Labirinto/fisiologia , Fibras Musgosas Hipocampais/fisiologia , Ratos , Ratos Wistar , Comportamento Espacial/fisiologiaRESUMO
BACKGROUND: The Gleason grading system is an important clinical practice for diagnosing prostate cancer in pathology images. However, this analysis results in significant variability among pathologists, hence creating possible negative clinical impacts. Artificial intelligence methods can be an important support for the pathologist, improving Gleason grade classifications. Consequently, our purpose is to construct and evaluate the potential of a Convolutional Neural Network (CNN) to classify Gleason patterns. METHODS: The methodology included 6982 image patches with cancer, extracted from radical prostatectomy specimens previously analyzed by an expert uropathologist. A CNN was constructed to accurately classify the corresponding Gleason. The evaluation was carried out by computing the corresponding 3 classes confusion matrix; thus, calculating the percentage of precision, sensitivity, and specificity, as well as the overall accuracy. Additionally, k-fold three-way cross-validation was performed to enhance evaluation, allowing better interpretation and avoiding possible bias. RESULTS: The overall accuracy reached 98% for the training and validation stage, and 94% for the test phase. Considering the test samples, the true positive ratio between pathologist and computer method was 85%, 93%, and 96% for specific Gleason patterns. Finally, precision, sensitivity, and specificity reached values up to 97%. CONCLUSION: The CNN model presented and evaluated has shown high accuracy for specifically pattern neighbors and critical Gleason patterns. The outcomes are in line and complement others in the literature. The promising results surpassed current inter-pathologist congruence in classical reports, evidencing the potential of this novel technology in daily clinical aspects.