Informatics in Medicine Unlocked

Advances in the cloud technology for secured distributed data storage, modern techniques for machine learning (ML), and access to large populations through mobile apps provide a unique opportunity for the healthcare industry professionals in the areas of early screening and medical diagnostics for certain diseases. This research study demonstrates the potential of ML using convolutional neural networks (CNN) for medical diagnostics of skin melanoma. Specifically, a comparison is presented between a shallow CNN (3-layers) with Resnet50 (50-layers) to classify open datasets of skin melanoma images as malignant or benign. Various ML performance metrics such as accuracy, recall, precision and receiver operating characteristic (ROC) are presented to recommend a deep learning model for the mobile app. Also, a novel framework is proposed for the scalability and adoption of ML-based medical diagnostics by large masses as a mobile app running on data-secure cloud platform. Using the open datasets, it is shown that skin cancer can be accurately diagnosed with a mobile phone app while maintaining patient privacy and data security.

Skin cancer, also known as melanoma, is generally diagnosed visually from the dermoscopic images, which is a tedious and time-consuming task for the dermatologist. Such a visual assessment, via the naked eye for skin cancers, is a challenging and arduous due to different artifacts such as low contrast, various noise, presence of hair, fiber, and air bubbles, etc. This article proposes a robust and automatic framework for the Skin Lesion Classification (SLC), where we have integrated image augmentation, Deep Convolutional Neural Network (DCNN), and transfer learning. The proposed framework was trained and tested on publicly available IEEE International Symposium on Biomedical Imaging (ISBI)-2017 dataset. The obtained average area under the receiver operating characteristic curve (AUC), recall, precision, and F1-score are respectively 0.87, 0.73, 0.76, and 0.74 for the SLC. Our experimental studies for lesion classification demonstrate that the proposed approach can successfully distinguish skin cancer with a high degree of accuracy, which has the capability of skin lesion identification for melanoma recognition.

Recently, Convolutional neural networks (CNN) have shown a growth due to their ability of learning different level image representations that helps in image classification in different fields. These networks have been trained on millions of images, so they gained a powerful ability of extracting the rightful features from input images, which results in accurate classification. In this research, we investigate the effects of transfer learning based convolutional neural networks for the iris tumor malignancy identification as it is notoriously hard to distinguish an iris nevus from an iris tumor. Features are transferred from a CNN trained on a source task, i.e. ImageNet, to a target task, i.e. iris tumor datasets. We transfer features learned from AlexNet and VGG-16 that are trained on ImageNet, to classify three different iris images types which are: iris nevus unaffected, iris cysts, and iris melanocytic tumors. The employed pre-trained models are modified by replacing their feedforward neural network classifier, Softmax, by a support vector machine (SVM) that is expected to slightly boost their performance (AlexNet-SVM and VGG16-SVM). All employed models are trained (fine-tuned) on a 60% of the available large dataset of iris images in order to investigate their power of generalization when trained using large amount of data. The networks are also tested on 40% of the data. The best performance was achieved by the VGG16-SVM which scored a high accuracy of 96.27% and strong features extraction capability as compared to the other models. Experimentally, it was seen that adding SVM contributed in improving the network performance compared to original models which use a feedforward neural network classifier.

The escalating incidence of skin lesions, coupled with a scarcity of dermatologists and the intricate nature of diagnostic procedures, has resulted in prolonged waiting periods. Consequently, morbidity and mortality rates stemming from untreated cancerous skin lesions have witnessed an upward trend. To address this issue, we propose a skin lesion classification model that leverages the efficient net B7 Convolutional Neural Network (CNN) architecture, enabling early screening of skin lesions based on camera images. The model is trained on a diverse dataset encompassing eight distinct skin lesion classes: Basal Cell Carcinoma (BCC), Squamous Cell Carcinoma (SCC), Melanoma (MEL), Dysplastic Nevi (DN), Benign Keratosis-Like lesions (BKL), Melanocytic Nevi (NV), and an Other class. Through multiple iterations of data preprocessing, as well as comprehensive error analysis, the model achieves a remarkable accuracy rate of 87%.

AO_SCPLOWBSTRACTC_SCPLOWCOVID-19 is a severe global problem, and AI can play a significant role in preventing losses by monitoring and detecting infected persons in early-stage. This paper aims to propose a high-speed and accurate fully-automated method to detect COVID-19 from the patients CT scan images. We introduce a new dataset that contains 48260 CT scan images from 282 normal persons and 15589 images from 95 patients with COVID-19 infections. At the first stage, this system runs our proposed image processing algorithm to discard those CT images that inside the lung is not properly visible in them. This action helps to reduce the processing time and false detections. At the next stage, we introduce a novel method for increasing the classification accuracy of convolutional networks. We implemented our method using the ResNet50V2 network and a modified feature pyramid network alongside our designed architecture for classifying the selected CT images into COVlD-19 or normal with higher accuracy than other models. After running these two phases, the system determines the condition of the patient using a selected threshold. We are the first to evaluate our system in two different ways. In the single image classification stage, our model achieved 98.49% accuracy on more than 7996 test images. At the patient identification phase, the system correctly identified almost 234 of 245 patients with high speed. We also investigate the classified images with the Grad-CAM algorithm to indicate the area of infections in images and evaluate our model classification correctness.

Actinic Keratosis (AK) is a frequent dermatological diagnosis which contributes to a large proportion of routine dermatopathology. A current development in histopathology is in the digitization of specimens by creating whole slide images (WSI) with slide scanners. Deep Learning Models (DLM) have been introduced to radiology or pathology for image recognition but dermatopathology lacks available solutions. Building on previous work about skin pathologies, this paper proposes a DLM following the U-Net architecture to detect AK in histopathological samples. In total, 297 histopathological slides (269 with AK and 28 without AK) have been retrospectively selected. They were randomly assigned to training, validation, and testing groups. Performance was evaluated by conducting a Case Control Accuracy Study on three levels of granularity. The DLM model achieved an overall accuracy of 99.13% on the WSI level, 99.02% on the patch level and an intersection over union (IoU) of 83.88%. The proposed DLM reliably recognizes AK in histopathological images, supporting the implementation of DLMs in dermatopathology practice. Given existing technical capabilities and advancements, DLMs could have a significant influence on dermatopathology routine in the future.

Artificial intelligence can be a key tool in the context of assisting in the diagnosis of dermatological conditions, particularly when performed by general practitioners with limited or no access to high resolution optical equipment. This study evaluates the performance of deep convolutional neural networks (DNNs) in the classification of seven pigmented skin lesions. Additionally, it assesses the improvement ratio in the classification performance when utilized by general practitioners. Open-source skin images were downloaded from the ISIC archive. Different DNNs (n=8) were trained based on a random dataset constituted by 8,015 images. A test set of 2,003 images has been used to assess the classifiers performance at low (300 x 224 RGB) and high (600 x 450 RGB) image resolution and aggregated clinical data (age, sex and lesion localization). We have also organized two different contests to compare the DNNs performance to that of general practitioners by means of unassisted image observation. Both at low and high image resolution, the DNNs framework being trained differentiated dermatological images with appreciable performance. In all cases, accuracy has been improved when adding clinical data to the framework. Finally, the lowest accurate DNN outperformed general practitioners. Physicians accuracy was statistically improved when allowed to use the output of this algorithmic framework as guidance. DNNS are proven to be high performers as skin lesion classifiers. The aim is to include these AI tools in the context of general practitioners whilst improving their diagnosis accuracy in a routine clinical scenario when or where the use of high-resolution equipment is not accessible.

ImportanceState-of-the art performance is achieved with a deep learning object detection model for acne detection. There is little current research on object detection in dermatology and acne in particular. As such, this work is early in this field and achieves state of the art performance. ObjectiveTrain an object detection model on a publicly available data set of acne photos. Design, Setting, and ParticipantsA deep learning model is trained with cross validation on a data set of facial acne photos. Main Outcomes and MeasuresObject detection models for detecting acne for single-class (acne) and multi-class (four severity levels). We train and evaluate the models using standard metrics such as mean average precision (mAP). Then we manually evaluate the model predictions on the test set, and calculate accuracy in terms of precision, recall, F1, true and false positive and negative detections. ResultsWe achieve state-of-the art mean average precision mAP@0.5 value of 37.97 for the single class acne detection task, and 26.50 for the 4-class acne detection task. Moreover, our manual evaluation shows that the single class detection model performs well on the validation set, achieving true positive 93.59 %, precision 96.45 % and recall 94.73 %. Conclusions and RelevanceWe are able to train a high-accuracy acne detection model using only a small publicly available data set of facial acne. Transfer learning on the pre-trained deep learning model yields good accuracy and high degree of transferability to patient submitted photographs. We also note that the training of standard architecture object detection models has given significantly better accuracy than more intricate and bespoke neural network architectures in the existing research literature. Key PointsO_ST_ABSQuestionC_ST_ABSCan deep learning-based acne detection models trained on a small data set of publicly available photos of patients with acne achieve high prediction accuracy? FindingsWe find that it is possible to train a reasonably good object detection model on a small, annotated data set of acne photos using standard deep learning architectures. MeaningDeep learning-based object detection models for acne detection can be a useful decision support tools for dermatologists treating acne patients in a digital clinical practice. It can prove a particularly useful tool for monitoring the time evolution of the acne disease state over prolonged time during follow-ups, as the model predictions give a quantifiable and comparable output for photographs over time. This is particularly helpful in teledermatological consultations, as a prediction model can be integrated in the patient-doctor remote communication.

COVID-19 or novel coronavirus disease, which has already been declared as a worldwide pandemic, at first had an outbreak in a small town of China, named Wuhan. More than two hundred countries around the world have already been affected by this severe virus as it spreads by human interaction. Moreover, the symptoms of novel coronavirus are quite similar to the general flu. Screening of infected patients is considered as a critical step in the fight against COVID-19. Therefore, it is highly relevant to recognize positive cases as early as possible to avoid further spreading of this epidemic. However, there are several methods to detect COVID-19 positive patients, which are typically performed based on respiratory samples and among them one of the critical approach which is treated as radiology imaging or X-Ray imaging. Recent findings from X-Ray imaging techniques suggest that such images contain relevant information about the SARS-CoV-2 virus. In this article, we have introduced a Deep Neural Network (DNN) based Faster Regions with Convolutional Neural Networks (Faster R-CNN) framework to detect COVID-19 patients from chest X-Ray images using available open-source dataset. Our proposed approach provides a classification accuracy of 97.36%, 97.65% of sensitivity, and a precision of 99.28%. Therefore, we believe this proposed method might be of assistance for health professionals to validate their initial assessment towards COVID-19 patients.

The purpose of this paper is to present a detailed investigation of the advantages of employing GraphLIME (Local Interpretable Model Explanations for Graph Neural Networks) for the trustworthy prediction of diabetes mellitus. Our pursuit involves identifying the strengths of GraphLIME combined with the attention-mechanism over the standard coupling of deep learning neural networks with the original LIME method. The system build this way, provided us a proficient method for extracting the most relevant features and applying the attention mechanism exclusively to those features. We have closely monitored the performance metrics of the two approaches and conducted a comparative analysis. Leveraging attention mechanisms, we have achieved an accuracy of 92.6% for the addressed problem. The models performance is meticulously demonstrated throughout the study, and the results are furthermore evaluated using the Receiver Operating Characteristic (ROC) curve. By implementing this technique on a dataset of 768 patients diagnosed with or without diabetes mellitus, we have successfully boosted the models performance by over 18%.

COVID-19, a severe acute respiratory syndrome aggressively spread among global populations in just a few months. Since then, it has had four dominant variants (Alpha, Beta, Gamma and Delta) that are far more contagious than original. Accurate and timely diagnosis of COVID-19 is critical for analysis of damage to lungs, treatment, as well as quarantine management [7]. CT, MRI or X-rays image analysis using deep learning provide an efficient and accurate diagnosis of COVID-19 that could help to counter its outbreak. With the aim to provide efficient multi-class COVID-19 detection, recently, COVID-19 Detection challenge using X-ray is organized [12]. In this paper, the late-fusion of features is extracted from pre-trained various convolutional neural networks and fine-tuned these models using the challenge dataset. The DensNet201 with Adam optimizer and EffecientNet-B3 are fine-tuned on the challenge dataset and ensembles the features to get the final prediction. Besides, we also considered the test time augmentation technique after the late-ensembling approach to further improve the performance of our proposed solution. Evaluation on Chest XR COVID-19 showed that our model achieved overall accuracy is 95.67%. We made the code is publicly available1. The proposed approach was ranked 6th in Chest XR COVID-19 detection Challenge [1].

COVID-19 is an infectious disease caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). As of December 2020, more than 72 million cases have been reported worldwide. The standard method of diagnosis is by Real-Time Reverse Transcription Polymerase Chain Reaction (rRT-PCR) from a Nasopharyngeal Swab. Currently, there is no vaccine or specific antiviral treatment for COVID-19. Due to rate of spreading of the disease manual detection among people is becoming more difficult because of a clear lack of testing capability. Thus there was need of a quick and reliable yet non-labour intensive detection technique. Considering that the virus predominantly appears in the form of a lung based abnormality I made use of Chest X-Rays as our primary mode of detection. For this detection system we made use of Posteroanterior (PA) Chest X-rays of people infected with Bacterial Pneumonia (2780 Images), Viral Pneumonia (1493 Images), Covid-19 (729 Images) as well as those of perfectly Healthy Individuals (1583 Images) procured from various Publicly Available Datasets and Radiological Societies. LungAI is a novel Convolutional Neural Network based on a Hybrid of the DarkNet and AlexNet architecture. The network was trained on 80% of the dataset with 20% kept for validation. The proposed Coronavirus Detection Model performed exceedingly well with an accuracy of 99.16%, along with a Sensitivity value of 99.31% and Specificity value of 99.14%. Thus LungAI has the potential to prove useful in managing the current Pandemic Situation by providing a reliable and fast alternative to Coronavirus Detection given strong results.

The early and rapid diagnosis of severe acute respiratory syndrome coronavirus 2 (SARS CoV-2), the main cause of fatal pandemic coronavirus disease 2019 (COVID-19), with the analysis of patients chest X-ray (CXR) images has life-saving importance for both patients and medical professionals. In this research a very simple novel and robust deep-learning convolutional neural network (CNN) model with less number of trainable-parameters is proposed to assist the radiologists and physicians in the early detection of COVID-19 patients. It also helps to classify patients into COVID-19, pneumonia and normal on the bases of analysis of augmented X-ray images. This augmented dataset contains 4803 COVID-19 from 686 publicly available chest X-ray images along with 5000 normal and 5000 pneumonia samples. These images are divided into 80% training and 20 % validation. The proposed CNN model is trained on training dataset and then tested on validation dataset. This model has a promising performance with a mean accuracy of 92.29%, precision of 99.96%, Specificity of 99.85% along with Sensitivity value of 85.92%. The result can further be improved if more data of expert radiologist is publically available.

Vitiligo, a dermatological condition characterized by depigmented patches on the skin, affects up to 2% of the global population. Its management is complex, often hindered by delayed diagnosis due to limited access to dermatologists and/ digital tools. Recent advancements in machine learning (ML) offer a potential solution by providing digital tools for early detection and management. This proof-of-concept study describes the development of a machine learning pipeline integrated into a mobile application for vitiligo assessment. Using a dataset of 1,309 images, including segmental and generalized vitiligo, the CNN was trained for binary classification with an accuracy of 95%. The model segments depigmented patches and conducts colorimetric analysis for precise evaluation. We compared traditional Woods lamp imaging with CNN-generated maps, showing comparable or superior results in detecting faint depigmentation. Developed using Flutter for cross-platform compatibility, the app enables patients to upload images for analysis and track disease progression. A Golang-based backend ensures robust data management, while a PostgreSQL database supports secure storage of patient information. The integration of Azure Active Directory enhances security and user authentication. This approach aims to bridge the gap in dermatological care by providing an accessible, ML-driven solution for vitiligo management. Future iterations will expand the applications capability to screen for other depigmentation disorders, incorporate automated scoring systems for more personalized patient management, and communication services.

Tuberculosis is an infectious disease that is leadingto the death of millions of people across the world. The mortalityrate of this disease is high in patients suffering from immuno-compromised disorders. The early diagnosis of this disease cansave lives and can avoid further complications. But the diagnosisof TB is a very complex task. The standard diagnostic tests stillrely on traditional procedures developed in the last century. Theseprocedures are slow and expensive. So this paper presents anautomatic approach for the diagnosis of TB from posteroanteriorchest x-rays. This is a two-step approach, where in the first stepthe lung regions are segmented from the chest x-rays using thegraph cut method, and then in the second step the transfer learn-ing of VGG16 combined with Bi-directional LSTM is used forextracting high-level discriminative features from the segmentedlung regions and then classification is performed using a fullyconnected layer. The proposed model is evaluated using data fromtwo publicly available databases namely Montgomery Countryset and Schezien set. The proposed model achieved accuracy andsensitivity of 97.76%, 97.01%and 96.42%, 94.11%on Schezienand Montgomery county datasets. This model enhanced thediagnostic accuracy of TB by 0.7%and 11.68%on Schezien andMontgomery county datasets.

The new coronavirus (known as COVID-19) was first identified in Wuhan and quickly spread worldwide, wreaking havoc on the economy and peoples everyday lives. Fever, cough, sore throat, headache, exhaustion, muscular aches, and difficulty breathing are all typical symptoms of COVID-19. A reliable detection technique is needed to identify affected individuals and care for them in the early stages of COVID-19 and reduce the viruss transmission. The most accessible method for COVID-19 identification is RT-PCR; however, due to its time commitment and false-negative results, alternative options must be sought. Indeed, compared to RT-PCR, chest CT scans and chest X-ray images provide superior results. Because of the scarcity and high cost of CT scan equipment, X-ray images are preferable for screening. In this paper, a pre-trained network, DenseNet169, was employed to extract features from X-ray images. Features were chosen by a feature selection method (ANOVA) to reduce computations and time complexity while overcoming the curse of dimensionality to improve predictive accuracy. Finally, selected features were classified by XGBoost. The ChestX-ray8 dataset, which was employed to train and evaluate the proposed method. This method reached 98.72% accuracy for two-class classification (COVID-19, healthy) and 92% accuracy for three-class classification (COVID-19, healthy, pneumonia).

The novel coronavirus (COVID-19) pandemic is pressurizing the healthcare systems across the globe and few of them are on the verge of failing. The detection of this virus as early as possible will help in contaminating the spread of it as the virus is mutating itself as fast as possible and currently there are about 4,300 strains of the virus according to the reports. Clinical studies have shown that most of the COVID-19 patients suffer from a lung infection similar to influenza. So, it is possible to diagnose lung infection using imaging techniques. Although a chest computed tomography (CT) scan has been shown to be an effective imaging technique for lung-related disease diagnosis, chest X-ray is more widely available across the hospitals due to its considerably lower cost and faster imaging time than CT scan. The advancements in the area of machine learning and pattern recognition has resulted in intelligent systems that analyze CT Scans or X-ray images and classify between pneumonia and normal patients. This paper proposes KE Sieve Neural Network architecture, which helps in the rapid diagnosis of COVID-19 using chest X-ray images. This architecture is achieving an accuracy of 98.49%. This noninvasive prediction method can assist the doctors in this pandemic and reduce the stress on health care systems.

Globally the devastating consequence of COVID-19 or Severe Acute Respiratory Syndrome-Coronavirus (SARS-CoV-2) has posed danger on the life of living beings. Doctors and scientists throughout the world are working day and night to combat the proliferation or transmission of this deadly disease in terms of technology, finances, data repositories, protective equipment, and many other services. Rapid and efficient detection of COVID-19 reduces the rate of spreading this deadly disease and early treatment improve the recovery rate. In this paper, we proposed a new framework to exploit powerful features extracted from the autoencoder and Gray Level Co-occurence Matrix (GLCM), combined with random forest algorithm for the efficient and fast detection of COVID-19 using computed tomographic images. The models performance is evident from its 97.78% accuracy, 96.78% recall, and 98.77% specificity.

The pandemic, originated by novel coronavirus 2019 (COVID-19), continuing its devastating effect on the health, well-being, and economy of the global population. A critical step to restrain this pandemic is the early detection of COVID-19 in the human body, to constraint the exposure and control the spread of the virus. Chest X-Rays are one of the non-invasive tools to detect this disease as the manual PCR diagnosis process is quite tedious and time-consuming. In this work, we propose an automated COVID-19 classifier, utilizing available COVID and non-COVID X-Ray datasets, along with High Resolution Network (HRNet) for feature extraction embedding with the UNet for segmentation purposes. To evaluate the proposed dataset, several baseline experiments have been performed employing numerous deep learning architectures. With extensive experiment, we got 99.26% accuracy, 98.53% sensitivity, and 98.82% specificity with HRNet which surpasses the performances of the existing models. Our proposed methodology ensures unbiased high accuracy, which increases the probability of incorporating X-Ray images into the diagnosis of the disease.

According to the World Health Organization (WHO), the coronavirus (COVID-19) pandemic is putting even the best healthcare systems across the world under tremendous pressure. The early detection of this type of virus will help in relieving the pressure of the healthcare systems. Chest X-rays has been playing a crucial role in the diagnosis of diseases like Pneumonia. As COVID-19 is a type of influenza, it is possible to diagnose using this imaging technique. With rapid development in the area of Machine Learning (ML) and Deep learning, there had been intelligent systems to classify between Pneumonia and Normal patients. This paper proposes the machine learning-based classification of the extracted deep feature using ResNet152 with COVID-19 and Pneumonia patients on chest X-ray images. SMOTE is used for balancing the imbalanced data points of COVID-19 and Normal patients. This non-invasive and early prediction of novel coronavirus (COVID-19) by analyzing chest X-rays can further be used to predict the spread of the virus in asymptomatic patients. The model is achieving an accuracy of 0.973 on Random Forest and 0.977 using XGBoost predictive classifiers. The establishment of such an approach will be useful to predict the outbreak early, which in turn can aid to control it effectively.