From complex datauncertain evidencestrict requirementsnoisy signalscomplex data to dependable systems.

Two first-author studies.Two very different signals.

1. 01IEEE Access · 2021Assessing the Impact of Segmentation on Wheat Stripe Rust Disease Classification Using Computer Vision and Deep LearningMeasured how image segmentation changes crop-disease classification performance.
   Quick look & citations

   Evaluates Watershed, GrabCut, and U2-Net segmentation against ResNet-18 classification, demonstrating how segmentation quality directly bounds field disease accuracy (highest accuracy 96.196% via U2-Net).

   Contribution

   Measured how image segmentation changes wheat stripe-rust classification performance.

   Method

   Compared classification workflows with different segmentation conditions within the study evaluation.

   Boundary

   A research result within the study dataset and protocol, not a claim of universal field performance.

   Full author abstract

   Wheat is a staple crop that is grown across the world due to its substantial contribution to human nutrition. Its significance is evident as it provides almost 20% of calories and protein required for daily human consumption. However, wheat yield is affected by rust disease that can reduce 30% of wheat production which is a serious threat to food security. In order to minimize the loss, it is crucial to identify precisely and localize the wheat rust disease and its infection types. For this purpose, several classification and segmentation techniques are used which are based on machine/deep learning models. This paper provides a realistic analysis and evaluation of various segmentation techniques including Watershed, Grab Cut, and U2-Net. These techniques are applied to the wheat stripe rust data to generate multiple datasets such as Watershed segmented data, GrabCut segmented data, and U2-Net segmented data. Subsequently, a pre-trained deep learning model, ResNet-18 is applied to these datasets to assess the impact of segmentation on classification accuracy. The highest classification accuracy (96.196%) is achieved on the dataset segmented by U2-Net. This research collates several state-of-the-art segmentation techniques in terms of correctness and their direct impact on classification accuracy which gives a pragmatic analysis for researchers to choose optimal segmentation technique. The research primarily focuses on the direct impact of segmentation on classification accuracy of wheat stripe rust, which has not been given sufficient focus in earlier researches.

   Copy BibTeXCopy APA
2. 0228th International Conference on Automation and Computing · 2023Studying the Impact of Varying Sample Length of ECG Signal on Classification AccuracyInvestigated how the duration of an ECG sample affects classification accuracy.
   Quick look & citations

   Evaluates Decision Tree, Random Forest, and SVM classifiers across Autonomic Aging, PTB-XL, and PTB Diagnostic datasets to show how window length fundamentally alters disease prediction accuracy.

   Contribution

   Investigated how ECG sample duration changes classification accuracy.

   Method

   Compared varying sample lengths under the paper’s experimental classification protocol.

   Boundary

   A study of signal duration and model evaluation, not clinical validation or diagnostic guidance.

   Full author abstract

   ECG Signals in digital format are being used for gathering various vital information and for predicting diseases in patients using advanced artificial intelligence and digital signal processing technologies. It is important to process the data sufficiently to get the optimal results. In addition to cleaning the dataset, it is important to select the right signal length. This study uses three publicly available datasets (namely autonomic aging dataset, PTB-XL dataset, and PTB Diagnostic dataset) to study the impact of varying sample lengths on classification accuracy i.e., understanding whether varying sample lengths can impact the accuracy and if so how different sample lengths for certain classification problems can affect it. The experiments performed used three machine learning algorithms which are Decision Tree Classifier, Random Forest Classifier, and State Vector Machine Classifier, and it was found that the signal length does impact the accuracy when predicting diseases using an ECG signal, and the behavior can vary depending on the type of data being used and the nature of classification problem, even if the same model is used.

   Copy BibTeXCopy APA