Aims: In response to the George B. Moody PhysioNet Challenge 2024, our team has undertaken developing algorithms for the digitization and classification of electrocardiograms (ECGs) captured from images or paper printouts. Our primary aim is to advance the accessibility and accuracy of ECG-based diagnoses, particularly focusing on underrepresented and underserved populations. Methods: Our team, buckeye_ai, has developed an approach leveraging machine learning techniques, particularly focusing on deep learning approaches. For the digitization aspect, we designed a model to extract features from ECG records, enhancing the conversion of analog signals into digital representations. Subsequently, for classification, we employed a vanilla Convolutional Neural Network (CNN) architecture, with two convolution and pooling layers along with two linear layers and a ReLU activation function, to categorize the ECG images into relevant classes. Additionally, we implemented preprocessing techniques to optimize the images for model input. Currently, our team is exploring the integration of Vision Transformers (ViTs) with Optical Character Recognition models to improve the digitization of paper ECGs. By utilizing the ability of ViTs to break down images into sequential patches, we aim to enhance the reconstruction of ECG signals from these images. Results: In the current challenge stage, our digitization model has a Reconstruction Signal-to-Noise Ratio (SNR) of -18.12. However, our current Classification F-measure stands at 0.14, indicating room for improvement. We are actively exploring strategies to enhance this metric and refine our approach. Conclusion: In conclusion, our participation in the George B. Moody PhysioNet Challenge 2024 has resulted in valuable insights and advancements in the field of ECG signal processing. Through our methodologies and different experimentation, we have made moves towards the digitization and classification of ECG images. Moving forward, we are committed to further tuning our models and exploring different techniques to improve the accuracy and efficiency of ECG signal analysis.