Search for: All records

In classification problems, mislabeled data can have a dramatic effect on the capability of a trained model. The traditional method of dealing with mislabeled data is through expert review. However, this is not always ideal, due to the large volume of data in many classification datasets, such as image datasets supporting deep learning models, and the limited availability of human experts for reviewing the data. Herein, we propose an ordered sample consensus (ORSAC) method to support data cleaning by flagging mislabeled data. This method is inspired by the random sample consensus (RANSAC) method for outlier detection. In short, the method involves iteratively training and testing a model on different splits of the dataset, recording misclassifications, and flagging data that is frequently misclassified as probably mislabeled. We evaluate the method by purposefully mislabeling subsets of data and assessing the method’s capability to find such data. We demonstrate with three datasets, a mosquito image dataset, CIFAR-10, and CIFAR-100, that this method is reliable in finding mislabeled data with a high degree of accuracy. Our experimental results indicate a high proficiency of our methodology in identifying mislabeled data across these diverse datasets, with performance assessed using different mislabeling frequencies. 

Mosquito-borne diseases continue to pose a great threat to global public health systems due to increased insecticide resistance and climate change. Accurate vector identification is crucial for effective control, yet it presents significant challenges. IDX - an automated computer vision-based device capable of capturing mosquito images and outputting mosquito species ID has been deployed globally resulting in algorithms currently capable of identifying 53 mosquito species. In this study, we evaluate deployed performance of the IDX mosquito species identification algorithms using data from partners in the Southeastern United States (SE US) and Papua New Guinea (PNG) in 2023 and 2024. This preliminary assessment indicates continued improvement of the IDX mosquito species identification algorithms over the study period for individual species as well as average regional accuracy with macro average recall improving from 55.3 % [Confidence Interval (CI) 48.9, 61.7] to 80.2 % [CI 77.3, 84.9] for SE US, and 84.1 % [CI 75.1, 93.1] to 93.6 % [CI 91.6, 95.6] for PNG using a CI of 90 %. This study underscores the importance of algorithm refinement and dataset expansion covering more species and regions to enhance identification systems thereby reducing the workload for human experts, addressing taxonomic expertise gaps, and improving vector control efforts.