Classifying Retinal Degeneration in Histological Sections Using Deep Learning.
Ontology highlight
ABSTRACT: Purpose
Artificial intelligence (AI) techniques are increasingly being used to classify retinal diseases. In this study we investigated the ability of a convolutional neural network (CNN) in categorizing histological images into different classes of retinal degeneration.Methods
Images were obtained from a chemically induced feline model of monocular retinal dystrophy and split into training and testing sets. The training set was graded for the level of retinal degeneration and used to train various CNN architectures. The testing set was evaluated through the best architecture and graded by six observers. Comparisons between model and observer classifications, and interobserver variability were measured. Finally, the effects of using less training images or images containing half the presentable context were investigated.Results
The best model gave weighted-F1 scores in the range 85% to 90%. Cohen kappa scores reached up to 0.86, indicating high agreement between the model and observers. Interobserver variability was consistent with the model-observer variability in the model's ability to match predictions with the observers. Image context restriction resulted in model performance reduction by up to 6% and at least one training set size resulted in a model performance reduction of 10% compared to the original size.Conclusions
Detecting the presence and severity of up to three classes of retinal degeneration in histological data can be reliably achieved with a deep learning classifier.Translational relevance
This work lays the foundations for future AI models which could aid in the evaluation of more intricate changes occurring in retinal degeneration, particularly in other types of clinically derived image data.
SUBMITTER: Al Mouiee D
PROVIDER: S-EPMC8196406 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:PurposeAccurate histological diagnosis in Hirschsprung disease (HD) is challenging, due to its complexity and potential for errors. In this study, we present an artificial intelligence (AI)-based method designed to identify ganglionic cells and hypertrophic nerves in HD histology.MethodsFormalin-fixed samples were used and an expert pathologist and a surgeon annotated these slides on a web-based platform, identifying ganglionic cells and nerves. Images were partitioned into square sections, augmented through data manipulation techniques and used to develop two distinct U-net models: one for detecting ganglionic cells and normal nerves; the other to recognise hypertrophic nerves.ResultsThe study included 108 annotated samples, resulting in 19,600 images after data augmentation and manually segmentation. Subsequently, 17,655 slides without target elements were excluded. The algorithm was trained using 1945 slides (930 for model 1 and 1015 for model 2) with 1556 slides used for training the supervised network and 389 for validation. The accuracy of model 1 was found to be 92.32%, while model 2 achieved an accuracy of 91.5%.ConclusionThe AI-based U-net technique demonstrates robustness in detecting ganglion cells and nerves in HD. The deep learning approach has the potential to standardise and streamline HD diagnosis, benefiting patients and aiding in training of pathologists.
| S-EPMC10687181 | biostudies-literature
Project description:The assessment of muscle condition is of great importance in various research areas. In particular, evaluating the degree of intramuscular fat (IMF) in tissue sections is a challenging task, which today is still mostly performed qualitatively or quantitatively by a highly subjective and error-prone manual analysis. We here realize the mission to make automated IMF analysis possible that (i) minimizes subjectivity, (ii) provides accurate and quantitative results quickly, and (iii) is cost-effective using standard hematoxylin and eosin (H&E) stained tissue sections. To address all these needs in a deep learning approach, we utilized the convolutional encoder-decoder network SegNet to train the specialized network IMFSegNet allowing to accurately quantify the spatial distribution of IMF in histological sections. Our fully automated analysis was validated on 17 H&E-stained muscle sections from individual sheep and compared to various state-of-the-art approaches. Not only does IMFSegNet outperform all other approaches, but this neural network also provides fully automated and highly accurate results utilizing the most cost-effective procedures of sample preparation and imaging. Furthermore, we shed light on the opacity of black-box approaches such as neural networks by applying an explainable artificial intelligence technique to clarify that the success of IMFSegNet actually lies in identifying the hard-to-detect IMF structures. Embedded in our open-source visual programming language JIPipe that does not require programming skills, it can be expected that IMFSegNet advances muscle condition assessment in basic research across multiple areas as well as in research fields focusing on translational clinical applications.
| S-EPMC10407270 | biostudies-literature
Project description:IntroductionEvaluating histopathology via machine learning has gained research and clinical interest, and the performance of supervised learning tasks has been described in various areas of medicine. Unsupervised learning of histological images has the advantage of reproducibility for labeling; however, the relationship between unsupervised evaluation and clinical information remains unclear in nephrology.MethodsWe propose an unsupervised approach combining convolutional neural networks (CNNs) and a visualization algorithm to cluster the histological images and calculate the score for patients. We applied the approach to the entire images or patched images of the glomerulus of kidney biopsy samples stained with hematoxylin and eosin obtained from 68 patients with immunoglobulin A nephropathy. We assessed the relationship between the obtained scores and clinical variables of urinary occult blood, urinary protein, serum creatinine (SCr), systolic blood pressure, and age.ResultsThe glomeruli of the patients were classified into 12 distinct classes and 10 patches. The output of the fine-tuned CNN, which we defined as the histological scores, had significant relationships with assessed clinical variables. In addition, the clustering and visualization results suggested that the defined clusters captured important findings when evaluating renal histopathology. For the score of the patch-based cluster containing crescentic glomeruli, SCr (coefficient = 0.09, P = 0.019) had a significant relationship.ConclusionThe proposed approach could successfully extract features that were related to the clinical variables from the kidney biopsy images along with the visualization for interpretability. The approach could aid in the quantified evaluation of renal histopathology.
| S-EPMC8418980 | biostudies-literature
Project description:Vocalizations are widely used for communication between animals. Mice use a large repertoire of ultrasonic vocalizations (USVs) in different social contexts. During social interaction recognizing the partner's sex is important, however, previous research remained inconclusive whether individual USVs contain this information. Using deep neural networks (DNNs) to classify the sex of the emitting mouse from the spectrogram we obtain unprecedented performance (77%, vs. SVM: 56%, Regression: 51%). Performance was even higher (85%) if the DNN could also use each mouse's individual properties during training, which may, however, be of limited practical value. Splitting estimation into two DNNs and using 24 extracted features per USV, spectrogram-to-features and features-to-sex (60%) failed to reach single-step performance. Extending the features by each USVs spectral line, frequency and time marginal in a semi-convolutional DNN resulted in a performance mid-way (64%). Analyzing the network structure suggests an increase in sparsity of activation and correlation with sex, specifically in the fully-connected layers. A detailed analysis of the USV structure, reveals a subset of male vocalizations characterized by a few acoustic features, while the majority of sex differences appear to rely on a complex combination of many features. The same network architecture was also able to achieve above-chance classification for cortexless mice, which were considered indistinguishable before. In summary, spectrotemporal differences between male and female USVs allow at least their partial classification, which enables sexual recognition between mice and automated attribution of USVs during analysis of social interactions.
| S-EPMC7347231 | biostudies-literature
Project description:BackgroundArtificial intelligence is being used for rehabilitation, including monitoring exercise compliance through sensor technology. AI classification of shoulder exercise wearing an IMU sensor has only been reported in normal (i.e. painless) subjects. To prove the feasibility of monitoring exercise compliance, we aimed to classify 11 types of shoulder rehabilitation exercises using an AI (artificial intelligence) algorithm in patients with shoulder pain. We had the patients wear an IMU-based sensor, collected data during exercise, and determined the accuracy of exercise classification.MethodsData were collected from 58 patients (27 males, 31 females, age range 37-82 years) diagnosed with shoulder diseases such as adhesive capsulitis and rotator cuff disease. 11 types of shoulder pain rehabilitation exercise programs were developed and repeated each exercise ten times per session while wearing an IMU sensor. The study applied the Rectified Linear Unit (ReLU) and the SoftMax as the activation function for hidden layers, the output layer.ResultsThe acquired data was used to train a DNN model using the multilayer perceptron algorithm. The trained model was used to classify 11 types of shoulder pain rehabilitation exercises. The training accuracy was 0.975 and the test accuracy was 0.925.ConclusionThe study demonstrates that IMU sensor data can effectively classify shoulder pain rehabilitation exercises, providing more appropriate feedback for patients. The model can be utilized to establish a system for remotely monitoring patients' exercise performance. The use of deep learning in patient monitoring and rehabilitation has significant potential to bring innovative changes to healthcare service delivery.
| S-EPMC10967210 | biostudies-literature
Project description:BackgroundHysteroscopy is a commonly used technique for diagnosing endometrial lesions. It is essential to develop an objective model to aid clinicians in lesion diagnosis, as each type of lesion has a distinct treatment, and judgments of hysteroscopists are relatively subjective. This study constructs a convolutional neural network model that can automatically classify endometrial lesions using hysteroscopic images as input.MethodsAll histopathologically confirmed endometrial lesion images were obtained from the Shengjing Hospital of China Medical University, including endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyps, and submucous myomas. The study included 1851 images from 454 patients. After the images were preprocessed (histogram equalization, addition of noise, rotations, and flips), a training set of 6478 images was input into a tuned VGGNet-16 model; 250 images were used as the test set to evaluate the model's performance. Thereafter, we compared the model's results with the diagnosis of gynecologists.ResultsThe overall accuracy of the VGGNet-16 model in classifying endometrial lesions is 80.8%. Its sensitivity to endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyp, and submucous myoma is 84.0%, 68.0%, 78.0%, 94.0%, and 80.0%, respectively; for these diagnoses, the model's specificity is 92.5%, 95.5%, 96.5%, 95.0%, and 96.5%, respectively. When classifying lesions as benign or as premalignant/malignant, the VGGNet-16 model's accuracy, sensitivity, and specificity are 90.8%, 83.0%, and 96.0%, respectively. The diagnostic performance of the VGGNet-16 model is slightly better than that of the three gynecologists in both classification tasks. With the aid of the model, the overall accuracy of the diagnosis of endometrial lesions by gynecologists can be improved.ConclusionsThe VGGNet-16 model performs well in classifying endometrial lesions from hysteroscopic images and can provide objective diagnostic evidence for hysteroscopists.
| S-EPMC7788977 | biostudies-literature
Project description:PurposeTo develop an automated diabetic retinopathy (DR) staging system using optical coherence tomography angiography (OCTA) images with a convolutional neural network (CNN) and to verify the feasibility of the system.MethodsIn this retrospective cross-sectional study, a total of 918 data sets of 3 × 3 mm2 OCTA images and 917 data sets of 6 × 6 mm2 OCTA images were obtained from 1118 eyes. A deep CNN and four traditional machine learning models were trained with annotations made by a retinal specialist based on ultra-widefield fluorescein angiography. Separately, the same images of the test data sets were independently graded by two human experts. The results of the CNN algorithm were compared with those of traditional machine learning-based classifiers and human experts.ResultsThe proposed CNN achieved an accuracy of 0.728, a sensitivity of 0.675, a specificity of 0.944, an F1 score of 0.683, and a quadratic weighted κ of 0.908 for a six-level staging task, which were far superior to the results of traditional machine learning methods or human experts. The CNN algorithm showed a better performance using 6 × 6 mm2 rather than 3 × 3 mm2 sized OCTA images and using combined data rather than a separate OCTA layer alone.ConclusionsCNN-based classification using OCTA images can provide reliable assistance to clinicians for DR classification.Translational relevanceThis CNN algorithm can guide the clinical decision for invasive angiography or referrals to ophthalmology specialists, helping to create more efficient diagnostic workflow in primary care settings.
| S-EPMC8899862 | biostudies-literature
Project description:BackgroundNowadays doctors and radiologists are overwhelmed with a huge amount of work. This led to the effort to design different Computer-Aided Diagnosis systems (CAD system), with the aim of accomplishing a faster and more accurate diagnosis. The current development of deep learning is a big opportunity for the development of new CADs. In this paper, we propose a novel architecture for a convolutional neural network (CNN) ensemble for classifying chest X-ray (CRX) images into four classes: viral Pneumonia, Tuberculosis, COVID-19, and Healthy. Although Computed tomography (CT) is the best way to detect and diagnoses pulmonary issues, CT is more expensive than CRX. Furthermore, CRX is commonly the first step in the diagnosis, so it's very important to be accurate in the early stages of diagnosis and treatment.ResultsWe applied the transfer learning technique and data augmentation to all CNNs for obtaining better performance. We have designed and evaluated two different CNN-ensembles: Stacking and Voting. This system is ready to be applied in a CAD system to automated diagnosis such a second or previous opinion before the doctors or radiology's. Our results show a great improvement, 99% accuracy of the Stacking Ensemble and 98% of accuracy for the the Voting Ensemble.ConclusionsTo minimize missclassifications, we included six different base CNN models in our architecture (VGG16, VGG19, InceptionV3, ResNet101V2, DenseNet121 and CheXnet) and it could be extended to any number as well as we expect extend the number of diseases to detected. The proposed method has been validated using a large dataset created by mixing several public datasets with different image sizes and quality. As we demonstrate in the evaluation carried out, we reach better results and generalization compared with previous works. In addition, we make a first approach to explainable deep learning with the objective of providing professionals more information that may be valuable when evaluating CRXs.
| S-EPMC9568999 | biostudies-literature
Project description:BackgroundAutomated text classification has many important applications in the clinical setting; however, obtaining labelled data for training machine learning and deep learning models is often difficult and expensive. Active learning techniques may mitigate this challenge by reducing the amount of labelled data required to effectively train a model. In this study, we analyze the effectiveness of 11 active learning algorithms on classifying subsite and histology from cancer pathology reports using a Convolutional Neural Network as the text classification model.ResultsWe compare the performance of each active learning strategy using two differently sized datasets and two different classification tasks. Our results show that on all tasks and dataset sizes, all active learning strategies except diversity-sampling strategies outperformed random sampling, i.e., no active learning. On our large dataset (15K initial labelled samples, adding 15K additional labelled samples each iteration of active learning), there was no clear winner between the different active learning strategies. On our small dataset (1K initial labelled samples, adding 1K additional labelled samples each iteration of active learning), marginal and ratio uncertainty sampling performed better than all other active learning techniques. We found that compared to random sampling, active learning strongly helps performance on rare classes by focusing on underrepresented classes.ConclusionsActive learning can save annotation cost by helping human annotators efficiently and intelligently select which samples to label. Our results show that a dataset constructed using effective active learning techniques requires less than half the amount of labelled data to achieve the same performance as a dataset constructed using random sampling.
| S-EPMC7941989 | biostudies-literature
Project description:Deep learning may transform health care, but model development has largely been dependent on availability of advanced technical expertise. Herein we present the development of a deep learning model by clinicians without coding, which predicts reported sex from retinal fundus photographs. A model was trained on 84,743 retinal fundus photos from the UK Biobank dataset. External validation was performed on 252 fundus photos from a tertiary ophthalmic referral center. For internal validation, the area under the receiver operating characteristic curve (AUROC) of the code free deep learning (CFDL) model was 0.93. Sensitivity, specificity, positive predictive value (PPV) and accuracy (ACC) were 88.8%, 83.6%, 87.3% and 86.5%, and for external validation were 83.9%, 72.2%, 78.2% and 78.6% respectively. Clinicians are currently unaware of distinct retinal feature variations between males and females, highlighting the importance of model explainability for this task. The model performed significantly worse when foveal pathology was present in the external validation dataset, ACC: 69.4%, compared to 85.4% in healthy eyes, suggesting the fovea is a salient region for model performance OR (95% CI): 0.36 (0.19, 0.70) p = 0.0022. Automated machine learning (AutoML) may enable clinician-driven automated discovery of novel insights and disease biomarkers.
| S-EPMC8119673 | biostudies-literature