Effective Preprocessing Technique Research Articles

Pre-processing the Photoplethysmography Signals for Enhancing the Cardiovascular Diseases Detection for Wrist Pulse Analysis in Nadi Ayurveda

INTRODUCTION: In recent years, Photoplethysmography (PPG) signal has played a vital role in detecting Cardiovascular Diseases (CVDs) in case of wrist pulse analysis emulating the Nadi Ayurveda. The PPG signals acquired from the sensor measurement are severely distorted by various artifacts, which significantly impact the accuracy of disease detection and hamper the disease diagnosis process. OBJECTIVES: Removing the noises is essential before detecting CVDs from the signals and thus, developing a simple and effective noise reduction method for enhancing the PPG signal quality constitutes a challenging research problem, particularly with prominent artifacts. METHODS: This paper designs an effective pre-processing technique that improves denoising methods to enhance the PPG signal quality. The design of pre-processing technique contains two major phases: Primary denoising-based artifact removal and secondary denoising-based Premature Ventricular Contraction (PVC) detection and Power-Line Interference (PLI) noise removal. The primary denoising method involves coarse and fine-grained filtering. The coarse-grained filtering removes the major artifacts, such as Baseline Wander (BLW) and Motion Artifacts (MA), by developing the Two-Stage Adaptive Noise Canceller (TANC) method. The fine-grained filtering process utilizes a detrended filter to filter the refined signal obtained from the TANC method. For the signals filtered from the primary denoising method, the secondary denoising method targets to detect the PVC-induced PPG signals from the decomposed high-frequency signals and removes high-frequency noise, such as PLI from noisy signals, by adopting the Wavelet Transform (WT) method. RESULTS: During the signal reconstruction process in the WT method, the research work reconstructs the denoised PPG signals along with the PVC-induced PPG signals. The experimental results of the noise removal methodology illustrated significant improvements in PPG signal quality. CONCLUSION: The designed pre-processing technique effectively denoises PPG signals, leading to enhanced signal quality which can further aid in accurate disease detection.

An ensemble classification approach for cervical cancer prediction using behavioral risk factors

Cervical cancer is a significant public health concern among females worldwide. Despite being preventable, it remains a leading cause of mortality. Early detection is crucial for successful treatment and improved survival rates. This study proposes an ensemble Machine Learning (ML) classifier for efficient and accurate identification of cervical cancer using medical data. The proposed methodology involves preparing two datasets using effective preprocessing techniques, extracting essential features using the scikit-learn package, and developing an ensemble classifier based on Random Forest, Support Vector Machine, Gaussian Naïve Bayes, and Decision Tree classifier traits. Comparison with other state-of-the-art algorithms using several ML techniques, including support vector machine, decision tree, random forest, Naïve Bayes, logistic regression, CatBoost, and AdaBoost, demonstrates that the proposed ensemble classifier outperforms them significantly, achieving accuracies of 98.06% and 95.45% for Dataset 1 and Dataset 2, respectively. The proposed ensemble classifier outperforms current state-of-the-art algorithms by 1.50% and 6.67% for Dataset 1 and Dataset 2, respectively, highlighting its superior performance compared to existing methods. The study also utilizes a five-fold cross-validation technique to analyze the benefits and drawbacks of the proposed methodology for predicting cervical cancer using medical data. The Receiver Operating Characteristic (ROC) curves with corresponding Area Under the Curve (AUC) values are 0.95 for Dataset 1 and 0.97 for Dataset 2, indicating the overall performance of the classifiers in distinguishing between the classes. Additionally, we employed SHapley Additive exPlanations (SHAP) as an Explainable Artificial Intelligence (XAI) technique to visualize the classifier’s performance, providing insights into the important features contributing to cervical cancer identification. The results demonstrate that the proposed ensemble classifier can efficiently and accurately identify cervical cancer and potentially improve cervical cancer diagnosis and treatment.

Performance discrepancy mitigation in heart disease prediction for multisensory inter-datasets.

Heart disease is one of the primary causes of morbidity and death worldwide. Millions of people have had heart attacks every year, and only early-stage predictions can help to reduce the number. Researchers are working on designing and developing early-stage prediction systems using different advanced technologies, and machine learning (ML) is one of them. Almost all existing ML-based works consider the same dataset (intra-dataset) for the training and validation of their method. In particular, they do not consider inter-dataset performance checks, where different datasets are used in the training and testing phases. In inter-dataset setup, existing ML models show a poor performance named the inter-dataset discrepancy problem. This work focuses on mitigating the inter-dataset discrepancy problem by considering five available heart disease datasets and their combined form. All potential training and testing mode combinations are systematically executed to assess discrepancies before and after applying the proposed methods. Imbalance data handling using SMOTE-Tomek, feature selection using random forest (RF), and feature extraction using principle component analysis (PCA) with a long preprocessing pipeline are used to mitigate the inter-dataset discrepancy problem. The preprocessing pipeline builds on missing value handling using RF regression, log transformation, outlier removal, normalization, and data balancing that convert the datasets to more ML-centric. Support vector machine, K-nearest neighbors, decision tree, RF, eXtreme Gradient Boosting, Gaussian naive Bayes, logistic regression, and multilayer perceptron are used as classifiers. Experimental results show that feature selection and classification using RF produce better results than other combination strategies in both single- and inter-dataset setups. In certain configurations of individual datasets, RF demonstrates 100% accuracy and 96% accuracy during the feature selection phase in an inter-dataset setup, exhibiting commendable precision, recall, F1 score, specificity, and AUC score. The results indicate that an effective preprocessing technique has the potential to improve the performance of the ML model without necessitating the development of intricate prediction models. Addressing inter-dataset discrepancies introduces a novel research avenue, enabling the amalgamation of identical features from various datasets to construct a comprehensive global dataset within a specific domain.

Enhancing Sentiment Analysis Accuracy by Optimizing Hyperparameters of SVM and Logistic Regression Models

The Analysis of Sentiments expressed on Twitter is a widely practiced application of Natural Language Processing (NLP) and Artificial Intelligence (AI). This process involves examining tweets to determine the emotional tone conveyed within the message. AI-based approaches are employed in Twitter sentiment analysis, typically following these steps: Data Collection, Data Preprocessing, and Sentiment Analysis, where AI techniques like Support Vector Machines (SVM) and Logistic Regression are utilized to categorize tweets into positive, negative, or neutral sentiments. Twitter data is a valuable source of information, serving diverse purposes such as real-time updates, user feedback, brand monitoring, market research, digital marketing, and political analysis. The Twitter API (Application Programming Interface) provides developers with tools and functionalities to access and interact with Twitter data, including tweets, user profiles, and timelines, enabling a wide range of applications and services. However, Twitter sentiment analysis presents challenges such as handling sarcasm, irony, colloquial language, and coping with the sheer volume and rapid flow of Twitter data. Nevertheless, with effective preprocessing techniques and AI methods, Twitter sentiment analysis can yield valuable insights into public opinion on various topics.

Improving Remote Photoplethysmography Performance through Deep-Learning-Based Real-Time Skin Segmentation Network

In recent years, health-monitoring systems have become increasingly important in the medical and safety fields, including patient and driver monitoring. Remote photoplethysmography is an approach that captures blood flow changes due to cardiac activity by utilizing a camera to measure transmitted or reflected light through the skin, but it has limitations in its sensitivity to changes in illumination and motion. Moreover, remote photoplethysmography signals measured from nonskin regions are unreliable, leading to inaccurate remote photoplethysmography estimation. In this study, we propose Skin-SegNet, a network that minimizes noise factors and improves pulse signal quality through precise skin segmentation. Skin-SegNet separates skin pixels and nonskin pixels, as well as accessories such as glasses and hair, through training on facial structural elements and skin textures. Additionally, Skin-SegNet reduces model parameters using an information blocking decoder and spatial squeeze module, achieving a fast inference time of 15 ms on an Intel i9 CPU. For verification, we evaluated Skin-SegNet using the PURE dataset, which consists of heart rate measurements from various environments. When compared to other skin segmentation methods with similar inference speeds, Skin-SegNet demonstrated a mean absolute percentage error of 1.18%, showing an improvement of approximately 60% compared to the 4.48% error rate of the other methods. The result even exhibits better performance, with only 0.019 million parameters, in comparison to DeepLabV3+, which has 5.22 million model parameters. Consequently, Skin-SegNet is expected to be employed as an effective preprocessing technique for facilitating efficient remote photoplethysmography on low-spec computing devices.

Financial derivative features based integrated potential fishing zone (IPFZ) Future forecast

In India, around 7 million people depend on fishing for their livelihoods. They are assisted with a reliable and fast brief forecast for the areas of fish aggregations. Habitat mapping is critical in supporting strategic choices on fish usage and protection. In conjunction with techniques for machine learning, remote control has made comprehensive fish mapping on relevant scales possible. In machine learning, supervised algorithms are utilized to make forecasts from datasets, when data is accessible without relating output factors. In this research work, Ocean Surface Temperature (OST) and Satellite derived Chlorophyl material are the basic inputs to generating the information of Potential Fishing Zone (PFZ). The 16 features and additional financial derivative features are used for accurate future prediction of PFZ. The unwanted and missing data are removed using effective pre-processing techniques. Among the various methods available for forecasting nonlinear phenomena, the Neural Network is the best and the efficient method to get a forecast. Therefore, the Function Fitting Neural Network (FFNN) technique is mainly used to predicting the Integrated Potential Fishing Zone (IPFZ). The practical analyses are performed by analysing the 80% -20%, 60% -40% and future data in terms of various parameters. From the results, it is proved that the suggested FFNN achieved 90% of accuracy, where the existing neural network achieved 86% of accuracy by implementing with financial derivative features for the 80% -20% of available dataset.

Embedded real-time objects’ hardness classification for robotic grippers

Robotic grippers can be equipped with tactile sensing systems to extract information from a manipulated object. The real-time classification of the physical properties of a grasped object on resource-constrained devices requires efficient and effective pre-processing techniques and machine-learning (ML) algorithms. In this paper, we propose a tactile sensing system mounted on the Baxter robot for the hardness classification of objects. In particular, we pre-processed the raw data with low computational cost techniques, and we designed three ML algorithms to provide real-time, energy-efficient, and low-memory impact classification on a resource-constrained microcontroller. Results show that convolutional neural networks (CNNs) achieve the best accuracy (>98%), while the support vector machine (SVM) presents the lowest memory occupation (1576 bytes), inference time (<0.077ms), and energy consumption (<5.74μJ).

IFlipper: Label Flipping for Individual Fairness

As machine learning becomes prevalent, mitigating any unfairness present in the training data becomes critical. Among the various notions of fairness, this paper focuses on the well-known individual fairness, which states that similar individuals should be treated similarly. While individual fairness can be improved when training a model (in-processing), we contend that fixing the data before model training (pre-processing) is a more fundamental solution. In particular, we show that label flipping is an effective pre-processing technique for improving individual fairness. Our system iFlipper solves the optimization problem of minimally flipping labels given a limit to the individual fairness violations, where a violation occurs when two similar examples in the training data have different labels. We first prove that the problem is NP-hard. We then propose an approximate linear programming algorithm and provide theoretical guarantees on how close its result is to the optimal solution in terms of the number of label flips. We also propose techniques for making the linear programming solution more optimal without exceeding the violations limit. Experiments on real datasets show that iFlipper significantly outperforms other pre-processing baselines in terms of individual fairness and accuracy on unseen test sets. In addition, iFlipper can be combined with in-processing techniques for even better results.

Shuffled shepherd political optimization‐based deep learning method for credit card fraud detection

SummaryDigital transactions based on credit cards are gradually increasing concept due to expediency. The amount of fraudulent transactions has intensely enlarged in modern days, because of the fast development of e‐services, namely e‐finance, mobile payments, and e‐commerce as well as the promotion of credit cards. Criminal fraud behaviors and user's payment behaviors are frequently varying, thus performance improvement of the fraud identification method and its stability are more challenging processes. The Shuffled Shepherd Political Optimization‐based Deep Residual network (SSPO‐based DRN) scheme is established for credit card fraud identification in this research. The SSPO is developed by merging the Political Optimization (PO) and Shuffled Shepherd Optimization Algorithm (SSOA). The quantile normalization model is an effective preprocessing technique, which normalizes the data for effective detection. Moreover, fisher score and class information gain effectively select the required features. Data augmentation is employed for increasing the data size, thereby the detection performance is improved. The Deep Residual Network (DRN) is employed for credit card fraud recognition, which is trained by devised SSPO algorithm. The SSPO‐based DRN approach achieved enhanced performance with testing sensitivity of 0.9279, specificity of 0.9023, and accuracy of 0.9120.

Human retinal biometric recognition system based on multiple feature extraction

Retinal biometric trait is one of the safest and most secure biometric traits that can be used for personal identification and authentication. The poor contrast in captured retinal images initially requires the adoption of a preprocessing technique, especially when the vessel segmentation is required. As a result, an effective preprocessing technique is employed before the feature extraction process. The reliability of the segmentation depends on the consistency of the contrast across the images. Due to the complexity of the retinal vascular patterns, segmenting the vessels manually is a time-consuming task that also requires extensive training and abilities to distinguish the thinner blood vessel pixels from the background of the retinal images. We present a unique blood vessel segmentation technique based on Radial Chebyshev Moments (RCMs) shape descriptors. Moment-based shape descriptors are typically strong, concise, and simple to compute. They are also insensitive to the object’s scale, rotation, and translation. Furthermore, to eliminate the eye movement and to overcome the various lighting problems, the following multiple features are extracted in the feature extraction stage to accurately authenticate a person: statistical (gray-level) features, shape-based (RCM) features, and structural features, such as minutiae, bifurcation points, optical disc, and enclosed regions. The principal components analysis methodology is used to optimize the feature vector, and the target image is identified using the support vector machine classifier. This step helps to determine whether the test image belongs to the authorized person or not. RIDB and VARIA are the two standard retinal authentication databases and DRIVE and STARE are the two general retinal databases that are used for evaluating the performance of the system. The system yielded a high identification rate of 98%, 97.5%, 97.2% and 94.4% with average processing time of 0.987 s, 0.40 s, 1.262 s and 0.65 s, respectively. From the results, it is evident that the proposed method is more effective and efficient for authenticating individuals.

An Ensemble Model for Consumer Emotion Prediction Using EEG Signals for Neuromarketing Applications.

Traditional advertising techniques seek to govern the consumer's opinion toward a product, which may not reflect their actual behavior at the time of purchase. It is probable that advertisers misjudge consumer behavior because predicted opinions do not always correspond to consumers' actual purchase behaviors. Neuromarketing is the new paradigm of understanding customer buyer behavior and decision making, as well as the prediction of their gestures for product utilization through an unconscious process. Existing methods do not focus on effective preprocessing and classification techniques of electroencephalogram (EEG) signals, so in this study, an effective method for preprocessing and classification of EEG signals is proposed. The proposed method involves effective preprocessing of EEG signals by removing noise and a synthetic minority oversampling technique (SMOTE) to deal with the class imbalance problem. The dataset employed in this study is a publicly available neuromarketing dataset. Automated features were extracted by using a long short-term memory network (LSTM) and then concatenated with handcrafted features like power spectral density (PSD) and discrete wavelet transform (DWT) to create a complete feature set. The classification was done by using the proposed hybrid classifier that optimizes the weights of two machine learning classifiers and one deep learning classifier and classifies the data between like and dislike. The machine learning classifiers include the support vector machine (SVM), random forest (RF), and deep learning classifier (DNN). The proposed hybrid model outperforms other classifiers like RF, SVM, and DNN and achieves an accuracy of 96.89%. In the proposed method, accuracy, sensitivity, specificity, precision, and F1 score were computed to evaluate and compare the proposed method with recent state-of-the-art methods.

Correlation-Guided Updating Strategy for Feature Selection in Classification With Surrogate-Assisted Particle Swarm Optimization

Classification data are usually represented by many features, but not all of them are useful. Without domain knowledge, it is challenging to determine which features are useful. Feature selection is an effective preprocessing technique for enhancing the discriminating ability of data, but it is a difficult combinatorial optimization problem because of the challenges of the huge search space and complex interactions between features. Particle swarm optimization (PSO) has been successfully applied to feature selection due to its efficiency and easy implementation. However, most existing PSO-based feature selection methods still face the problem of falling into local optima. To solve this problem, this article proposes a novel PSO-based feature selection approach, which can continuously improve the quality of the population at each iteration. Specifically, a correlation-guided updating strategy based on the characteristic of data is developed, which can effectively use the information of the current population to generate more promising solutions. In addition, a particle selection strategy based on a surrogate technique is presented, which can efficiently select particles with better performance in both convergence and diversity to form a new population. Experimental comparing the proposed approach with a few state-of-the-art feature selection methods on 25 classification problems demonstrate that the proposed approach is able to select a smaller feature subset with higher classification accuracy in most cases.

Effective fuzzy joint mutual information feature selection based on uncertainty region for classification problem

Classification problem widely exists in real-world applications. Unfortunately, data quality is the main challenge of the classification models, especially when the data includes irrelevant and redundant features. Feature selection (FS) is an effective preprocessing technique to enhance the quality of the data. For this, an integration of information theory and fuzzy sets introduced powerful measures, such as fuzzy information measures, to develop many feature selection methods. However, estimating fuzzy information measures is not only costly in the space and runtime but also may be affected by the bias between the certainty and uncertainty regions. This paper proposes a novel instance selection based on uncertainty region (ISUR) to overcome these limitations. Then, a state-of-the-art FS method, called fuzzy joint mutual information (FJMI), has been adapted to design an effective FS method, called fuzzy joint mutual information feature selection based on uncertainty region (FJMIUR). The proposed method consists of two processes: instance selection and feature selection. The former selects the uncertainty region that improves the estimation of fuzzy information measures and reduces the consumed cost while the latter selects the most significant features. Using 20 real-world classification datasets, comparative experiments, including well-known and state-of-the-art FS methods, were conducted to evaluate the effectiveness of FJMIUR. The results show the outperformance of FJMIUR in most cases according to six classification measures, average percentage of selected features, space, and runtime.

Effective Preprocessing and Normalization Techniques for COVID-19 Twitter Streams with POS Tagging via Lightweight Hidden Markov Model

The major focus of this research work is to refine the basic preprocessing steps for the unstructured text content and retrieve the potential conceptual features for further enhancement processes such as semantic enrichment and named entity recognition. Although some of the preprocessing techniques such as text tokenization, normalization, and Part-of-Speech (POS) tagging work exceedingly well on formal text, it has not performed well when it is applied into informal text such as tweets and short messages. Hence, we have given the enhanced text normalization techniques to reduce the complexity persist over the twitter streams and eliminate the overfitting issues such as text anomalies and irregular boundaries while fixing the grammar of the text. The hidden Markov model (HMM) has been pervasively used to extract the core lexical features from the Twitter dataset and suitably adapt the external documents to supplement the extraction techniques to complement the tweet context. Using this Markov process, the POS tags are identified as states of the Markov process, and words are the desired results of the model. As this process is very crucial for the next stage of entity extraction and classification, the effective handling of informal text is considered to be important and therefore proposed the most effective hybrid approach to deal with the issues appropriately.

An approach for brain tumor detection using optimal feature selection and optimized deep belief network

Nowadays, a Magnetic Resonance Image (MRI) scan acts as an efficient tool for efficiently detecting the abnormal tissues present in the brain. It is a complex process for radiologists to diagnose as well as classify the tumor from several images. This paper develops an intelligent method for the accurate detection of brain tumors. Initially, the pre-processing is performed for the input MRI image using the skull stripping and the entropy-based trilateral filtering methods. Further, fuzzy centroid-based region growing is adopted for segmenting the tumor from the image. Once the tumor is segmented, feature extraction is done using four sets of well-performing features like Gray-Level Co-occurrence Matrix (GLCM), Gray Level Run-length Matrix (GLRM), statistical features, and shape features. The optimal feature selection is performed by the hybrid meta-heuristic algorithm termed Group Search-based Multi-Verse optimization (GS-MVO). Finally, the optimally selected features are given to a deep learning algorithm called Deep Belief Network (DBN). The weight is optimized by the same GS-MVO that classifies the final image as normal or abnormal. The simulation outcomes are performed by the standard benchmark database which proves that the developed technique obtains a high classification accuracy. From the analysis, the accuracy of the proposed GS-MVO-DBN is 9.09% superior to SVM, 7.14% superior to NN, 3.45% superior to DBN, 17.65% superior to CNN, 15.38% superior to NN-CNN, and 1.69% superior to COR-CSO-CNN-NN. The proposed GS-MVO-DBN is very effective in accurately detecting brain tumors. In the future, it is encouraged to work on challenging parts of the tumor region like edema, necrosis, and active regions with the help of the fusion process of multi-modality MRI images and effective pre-processing techniques incorporated with innovative deep learning methods.

Intelligent Information Retrieval for Reducing Missed Cancer and Improving the Healthcare System

This study presents an intelligent information retrieval system that will effectively extract useful information from breast cancer datasets and utilized that information to build a classification model. The proposed model will reduce the missed cancer rate by providing a comprehensive decision support to the radiologist. The model is built on two datasets, Wisconsin Breast Cancer Dataset (WBCD) and 365 free text mammography reports from a hospital. Effective pre-processing techniques including filling missing values with regression, an effective Natural Language Processing (NLP) Parser is developed to handle free text mammography reports, balancing the dataset with Synthetic Minority Oversampling (SMOTE) was applied to prepare the dataset for learning. Most relevant features were selected with the help of filter method and tf-idf scores. K-NN and SGD classifiers are optimized with optimum value of k for K-NN and hyper tuning the SGD parameters with grid search technique.

A deep learning approach using effective preprocessing techniques to detect COVID-19 from chest CT-scan and X-ray images

Coronavirus disease-19 (COVID-19) is a severe respiratory viral disease first reported in late 2019 that has spread worldwide. Although some wealthy countries have made significant progress in detecting and containing this disease, most underdeveloped countries are still struggling to identify COVID-19 cases in large populations. With the rising number of COVID-19 cases, there are often insufficient COVID-19 diagnostic kits and related resources in such countries. However, other basic diagnostic resources often do exist, which motivated us to develop Deep Learning models to assist clinicians and radiologists to provide prompt diagnostic support to the patients. In this study, we have developed a deep learning-based COVID-19 case detection model trained with a dataset consisting of chest CT scans and X-ray images. A modified ResNet50V2 architecture was employed as deep learning architecture in the proposed model. The dataset utilized to train the model was collected from various publicly available sources and included four class labels: confirmed COVID-19, normal controls and confirmed viral and bacterial pneumonia cases. The aggregated dataset was preprocessed through a sharpening filter before feeding the dataset into the proposed model. This model attained an accuracy of 96.452% for four-class cases (COVID-19/Normal/Bacterial pneumonia/Viral pneumonia), 97.242% for three-class cases (COVID-19/Normal/Bacterial pneumonia) and 98.954% for two-class cases (COVID-19/Viral pneumonia) using chest X-ray images. The model acquired a comprehensive accuracy of 99.012% for three-class cases (COVID-19/Normal/Community-acquired pneumonia) and 99.99% for two-class cases (Normal/COVID-19) using CT-scan images of the chest. This high accuracy presents a new and potentially important resource to enable radiologists to identify and rapidly diagnose COVID-19 cases with only basic but widely available equipment.

Nonlinear registration as an effective preprocessing technique for Deep learning based classification of disease.

A number of machine learning (ML), and particularly in recent years, deep learning (DL) approaches have been proposed for automatic classification of Alzheimer's disease (AD) using brain structural magnetic resonance imaging (MRI) data. However, the data available are limited in the case of this specific disease. Training a DL model with a large number of feature parameters on a small dataset of MRI scans will likely lead to overfitting. Overfitting reduces the generality and efficiency of the model. In this study, we show that a traditional nonlinear transformation from native space to template space, as a preprocessing stage, is effective in reducing overfitting through the reduction of spatial variations in the input data. To evaluate this effectiveness, we compare two different pre-processing approaches for DL-based AD classification task: (1) affine registration and (2) nonlinear diffeomorphic anatomical registration using exponentiated Lie algebra (DARTEL). The results show that the accuracy of the nonlinear registration based approach is much higher than the affine registration based approach. Furthermore, from the classification results obtained with noisy images, DARTEL is less susceptible to noise than affine registration. In summary, our experimental results suggest that nonlinear transformation is a preferable preprocessing step for training DL-based AD classification models on limited size datasets.

Effective Image Pre-Processing Techniques with Deep Learning for Leukemia Detection

Effective Image Pre-Processing Techniques with Deep Learning for Leukemia Detection

ECG Paper Record Digitization and Diagnosis Using Deep Learning.

PurposeElectrocardiogram (ECG) is one of the most essential tools for detecting heart problems. Till today most of the ECG records are available in paper form. It can be challenging and time-consuming to manually assess the ECG paper records. Hence, automated diagnosis and analysis are possible if we digitize such paper ECG records.MethodsThe proposed work aims to convert ECG paper records into a 1-D signal and generate an accurate diagnosis of heart-related problems using deep learning. Camera-captured ECG images or scanned ECG paper records are used for the proposed work. Effective pre-processing techniques are used for the removal of shadow from the images. A deep learning model is used to get a threshold value that separates ECG signal from its background and after applying various image processing techniques threshold ECG image gets converted into digital ECG. These digitized 1-D ECG signals are then passed to another deep learning model for the automated diagnosis of heart diseases into different classes such as ST-segment elevation myocardial infarction (STEMI), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), and T-wave abnormality.ResultsThe accuracy of deep learning-based binarization is 97%. Further deep learning-based diagnosis approach of such digitized paper ECG records was having an accuracy of 94.4%.ConclusionsThe digitized ECG signals can be useful to various research organizations because the trends in heart problems can be determined and diagnosed from preserved paper ECG records. This approach can be easily implemented in areas where such expertise is not available.Supplementary InformationThe online version contains supplementary material available at 10.1007/s40846-021-00632-0.