Deep Lungs: Revolutionizing Cancer Detection with Neural Networks

Cardio-respiratory disorders account for approximately 30% of total mortality in India, with their incidence steadily rising due to factors such as increasing urban pollution, sedentary lifestyles, and delayed clinical diagnosis, as highlighted by the Indian Council of Medical Research. Conventional auscultation using stethoscopes is highly dependent on the clinician’s expertise and experience, which can lead to variability and potential inaccuracies in diagnosis. To address these limitations, this study presents an automated cardio-respiratory sound classification framework designed to enhance early detection and diagnostic consistency through machine learning. The proposed system incorporates a user-friendly Graphical User Interface (GUI) with role-based functionality, allowing administrators to manage model training while enabling end-users to perform real-time predictions. Audio signals are preprocessed using the Librosa library to extract informative acoustic features such as Mel-Frequency Cepstral Coefficients (MFCC), Chroma features, and Mel Spectrogram representations, ensuring comprehensive characterization of both heart and lung sounds. A range of machine learning models, including Quadratic Discriminant Analysis (QDA), Gradient Boosting Classifier (GBC), Naïve Bayes Classifier (NBC), and Logistic Regression Classifier (LRC), are developed and comparatively evaluated. Furthermore, a hybrid deep learning architecture combining a Bi-directional Convolutional Neural Network (BiCNN) with a Bi-directional Gated Recurrent Unit (BiGRU) is introduced to effectively capture both spatial (spectral) and temporal dependencies within the audio data. The system is capable of performing dual classification by identifying both heart sound types and lung sound types simultaneously. Its performance is rigorously assessed using standard evaluation metrics, including accuracy, precision, recall, F1-score, and Receiver Operating Characteristic–Area Under Curve (ROCAUC), demonstrating its potential as a reliable and efficient tool for clinical decision support.

The development of medical image analysis algorithm is a complex process including the multiple sub-steps of model training, data visualization, human–computer interaction and graphical user interface (GUI) construction. To accelerate the development process, algorithm developers need a software tool to assist with all the sub-steps so that they can focus on the core function implementation. Especially, for the development of deep learning (DL) algorithms, a software tool supporting training data annotation and GUI construction is highly desired. In this work, we constructed AnatomySketch, an extensible open-source software platform with a friendly GUI and a flexible plugin interface for integrating user-developed algorithm modules. Through the plugin interface, algorithm developers can quickly create a GUI-based software prototype for clinical validation. AnatomySketch supports image annotation using the stylus and multi-touch screen. It also provides efficient tools to facilitate the collaboration between human experts and artificial intelligent (AI) algorithms. We demonstrate four exemplar applications including customized MRI image diagnosis, interactive lung lobe segmentation, human-AI collaborated spine disc segmentation and Annotation-by-iterative-Deep-Learning (AID) for DL model training. Using AnatomySketch, the gap between laboratory prototyping and clinical testing is bridged and the development of MIA algorithms is accelerated. The software is opened at https://github.com/DlutMedimgGroup/AnatomySketch-Software.

Response

1 - Congruence of deep learning in biomedical engineering: future prospects and challenges

Urban areas face increasing challenges in constructing resilient infrastructure on weak or unstable soils, especially amid the impacts of climate change and rapid urbanization. This study introduces an innovative AI-driven framework that integrates advanced hybrid deep learning architectures namely Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Transformer models for accurately predicting the Unconfined Compressive Strength (UCS) of Nano-silica (NS) stabilized soils. A key innovation of this research lies in the development of a novel CNN-Transformer hybrid model, which outperforms traditional and standalone AI models, achieving an R2 of 0.97 and RMSE of 0.22 reducing prediction error by over 49%. Robustness was further validated using a 10,000-iteration Monte Carlo simulation. In addition to predictive modeling, this study pioneers a comparative Life Cycle Assessment (LCA) between NS and cement-based stabilization, revealing that NS reduces CO₂ emissions by 55%, lowers energy consumption by 73%, and improves material efficiency. Furthermore, a user-friendly Graphical User Interface (GUI) tool has been developed, enabling real-time optimization of NS dosage for practical implementation in urban projects. This research not only contributes a high-performance predictive tool but also supports sustainable construction practices, aligning with SDG 9 (Industry, Innovation, and Infrastructure) and SDG 12 (Responsible Consumption and Production).

A new Graphical User Interface (GUI) was developed using Laboratory Virtual Instrumentation Engineering Workbench (LabVIEW) for a high-resolution, high-sensitivity Solid State X-ray Image Intensifier (SSXII), which is a new x-ray detector for radiographic and fluoroscopic imaging, consisting of an array of Electron-Multiplying CCDs (EMCCDs) each having a variable on-chip electron-multiplication gain of up to 2000× to reduce the effect of readout noise. To enlarge the field-of-view (FOV), each EMCCD sensor is coupled to an x-ray phosphor through a fiberoptic taper. Two EMCCD camera modules are used in our prototype to form a computer-controlled array; however, larger arrays are under development. The new GUI provides patient registration, EMCCD module control, image acquisition, and patient image review. Images from the array are stitched into a 2k×1k pixel image that can be acquired and saved at a rate of 17 Hz (faster with pixel binning). When reviewing the patient's data, the operator can select images from the patient's directory tree listed by the GUI and cycle through the images using a slider bar. Commonly used camera parameters including exposure time, trigger mode, and individual EMCCD gain can be easily adjusted using the GUI. The GUI is designed to accommodate expansion of the EMCCD array to even larger FOVs with more modules. The high-resolution, high-sensitivity EMCCD modular-array SSXII imager with the new user-friendly GUI should enable angiographers and interventionalists to visualize smaller vessels and endovascular devices, helping them to make more accurate diagnoses and to perform more precise image-guided interventions.

The use of machine learning (ML) has been increasing rapidly in the medical imaging field, including computer-aided diagnosis (CAD), radiomics, and medical image analysis. Recently, an ML area called deep learning emerged in the computer vision field and became very popular in many fields. It started from an event in late 2012, when a deep-learning approach based on a convolutional neural network (CNN) won an overwhelming victory in the best-known worldwide computer vision competition, ImageNet Classification. Since then, researchers in virtually all fields, including medical imaging, have started actively participating in the explosively growing field of deep learning. In this paper, the area of deep learning in medical imaging is overviewed, including (1) what was changed in machine learning before and after the introduction of deep learning, (2) what is the source of the power of deep learning, (3) two major deep-learning models: a massive-training artificial neural network (MTANN) and a convolutional neural network (CNN), (4) similarities and differences between the two models, and (5) their applications to medical imaging. This review shows that ML with feature input (or feature-based ML) was dominant before the introduction of deep learning, and that the major and essential difference between ML before and after deep learning is the learning of image data directly without object segmentation or feature extraction; thus, it is the source of the power of deep learning, although the depth of the model is an important attribute. The class of ML with image input (or image-based ML) including deep learning has a long history, but recently gained popularity due to the use of the new terminology, deep learning. There are two major models in this class of ML in medical imaging, MTANN and CNN, which have similarities as well as several differences. In our experience, MTANNs were substantially more efficient in their development, had a higher performance, and required a lesser number of training cases than did CNNs. "Deep learning", or ML with image input, in medical imaging is an explosively growing, promising field. It is expected that ML with image input will be the mainstream area in the field of medical imaging in the next few decades.

This paper explores how well deep learning models trained on chest CT images can diagnose COVID-19 infected people in a fast and automated process. To this end, we adopted advanced deep network architectures and proposed a transfer learning strategy using custom-sized input tailored for each deep architecture to achieve the best performance. We conducted extensive sets of experiments on two CT image datasets, namely, the SARS-CoV-2 CT-scan and the COVID19-CT. The results show superior performances for our models compared with previous studies. Our best models achieved average accuracy, precision, sensitivity, specificity, and F1-score values of , , , , and on the SARS-CoV-2 dataset, and , , , , and on the COVID19-CT dataset, respectively. For better interpretability of the results, we applied visualization techniques to provide visual explanations for the models’ predictions. Feature visualizations of the learned features show well-separated clusters representing CT images of COVID-19 and non-COVID-19 cases. Moreover, the visualizations indicate that our models are not only capable of identifying COVID-19 cases but also provide accurate localization of the COVID-19-associated regions, as indicated by well-trained radiologists.

In this paper, a Graphical User Interface (GUI) is designed and developed using Python programming language for detection of forest fires at early stages, which is efficient, cost-effective, flexible, secure and uses open-source technologies. The user-friendly Graphical User Interface is interfaced directly with the computer port to fetch the data with the help of Python modules, performs an analysis on the data and predict whether there is a situation of fire or not and displays the status on the Graphical User Interface. Data is also stored in an Excel file for further analysis. The possibility to detect forest fire situation at its early stages makes this GUI a valuable software in environmental disasters and research applications. Numerical results along with future work have been discussed in this paper.

The SARS-CoV-2 (COVID-19) has propagated rapidly around the world, and it became a global pandemic. It has generated a catastrophic effect on public health. Thus, it is crucial to discover positive cases as early as possible to treat touched patients fastly. Chest CT is one of the methods that play a significant role in diagnosing 2019-nCoV acute respiratory disease. The implementation of advanced deep learning techniques combined with radiological imaging can be helpful for the precise detection of the novel coronavirus. It can also be assistive to surmount the difficult situation of the lack of medical skills and specialized doctors in remote regions. This paper presented Deep Transfer Learning Pipelines with Apache Spark and KerasTensorFlow combined with the Logistic Regression algorithm for automatic COVID-19 detection in chest CT images, using Convolutional Neural Network (CNN) based models VGG16, VGG19, and Xception. Our model produced a classification accuracy of 85.64, 84.25, and 82.87 %, respectively, for VGG16, VGG19, and Xception. HIGHLIGHTS Deep Transfer Learning Pipelines with Apache Spark and Keras TensorFlow combined with Logistic Regression using CT images to screen for Corona Virus Disease (COVID-19) Automatic detection of COVID-19 in chest CT images Convolutional Neural Network (CNN) based models VGG16, VGG19, and Xception to predict COVID-19 in Computed Tomography image GRAPHICAL ABSTRACT

Deep learning is a fast flattering highest development in the machine learning technique leading to augment performance in diverse medical applications. The tributary artificial intelligence is rapidly becoming metamorphic in the domain of Healthcare Technology rendering facility to analyze the medical data with high rapidity and exactness. Due to the overwhelming development of artificial intelligence (AI), the contemporary encroachments of deep learning employing advanced and futuristic deep learning methods has become a vivacious research domain for processing medical data in the health care industries. This paper presents a complete analysis about the deep structured learning based methods used in medical image analysis and also elucidate about the advancement of convolutional neural network which is a major class of deep neural networks used in medical image processing. It also gives insight about the fundamentals of deep learning methods and reviews in image interpretation. It also explains about the implementation of an integrated framework for medical image synthesis combining Speckle GAN and Di-Conv-AE-Net.

The emergence of deep learning has paved to solve many problems in the real world. COVID-19 pandemic, since the late 2019, has been affecting lives of people across the globe. Chest CT scan images are used to detect it and know its severity in patients. The problem with many existing solutions in COVID-19 detection using CT scan images is that inability to detect the infection when it is in initial stages. As the infection can exist on varied scales, there is need for more comprehensive approach that can ascertain the disease at all scales. Towards this end, we proposed a deep learning-based framework known as Automated Deep Learning-based COVID-19 Detection Framework (ADL-CDF). It does not need a human medical expert in diagnosis as it is capable of detecting automatically. The framework is assisted by two algorithms that involve image processing and deep learning. The first algorithm known as Region of Interest (ROI)-based Image Filtering (ROI-IF) which analyses given input CT scan images of a patient and discards the ones where ROI is missing. This algorithm minimizes time taken for processing besides reducing false positive rate. The second algorithm is known as Multi-Scale Feature Selection algorithm that fits into the deep learning framework’s pipeline to leverage detection performance of the ADL-CDF. The proposed framework is evaluated against ResNet50V2 and Xception. Our empirical study revealed that our model outperforms the state of the art.

Rationale. The complex interplay of factors contributing to the pathobiology of obesity-associated asthma is poorly understood. Advancing understanding of the unique disease mechanisms driving gas exchange abnormalities in asthma with comorbid obesity may improve patient management in this difficult to treat population. Hyperpolarized 129Xe MRI has revealed increased ventilation defect percent (VDP) in asthma, and, separately, increased membrane high percent (MHP) in obese healthy participants. However, XeMRI has not been studied specifically in obesity-associated asthma. Here, we characterized patterns of XeMRI gas exchange in obese asthma and assess correlations with clinical measures of lung function, including impulse oscillometry. Methods. N=23 asthma participants (18F, 47±9 years) and 7 healthy controls (5F, 46±12 years) underwent supine chest CT and XeMRI gas exchange imaging scans, spirometry, and impulse oscillometry. All subjects had BMI>30 kg/m2. Chest CT scans were analyzed with the VIDA software package to delineate segmental boundaries and registered to the XeMRI images, with bronchopulmonary segments broadly categorized as either “anterior” or “posterior”. Differences in VDP, MHP, and red blood cell defect percent (RDP) in asthma vs. controls were assessed using the Wilcoxon rank-sum test. Correlations between VDP and forced expiratory volume in 1 second (FEV1), resistance at 5Hz (R5), change in resistance (R5-20), and area of reactance (AX) on impulse oscillometry, and methacholine PC20 (MChPC20), were assessed using the Spearman correlation. Values of MHP in posterior vs. anterior segments was assessed using the Wilcoxon rank-sum test. Results. No significant differences were observed in VDP, MHP, or RDP in the asthma vs. control groups. Across all participants, VDP was correlated with AX (ρ=0.53, p<0.01) but not FEV1 (ρ=-0.31, p=0.11), and MHP was significantly higher in the posterior segments than the anterior segments (median [1Q-3Q] of 3.8% [0.9%-23.9%] vs. 0.6% [0%-7.6%] respectively, p<0.01). VDP was negatively correlated with MChPC20 (ρ=-0.56, p=0.03) in asthma participants. Conclusion. No significant differences were observed in XeMRI gas exchange function in obese asthma vs. obese control participants. VDP correlated with AX across all subjects, suggesting a link between peripheral airway compliance and VDP in obesity regardless of diagnosis. Further, the link between VDP and MChPC20 in subjects with obesity-associated asthma suggests airway hyperreactivity contributes to VDP, despite the lack of strong correlation with FEV1. Finally, increased MHP in posterior vs. anterior segments suggests gravitational dependence rather than regional inflammation or fibrosis. Ongoing work will assess regional gas exchange patterns pre- and post-bronchodilator in this complex population.

The escalating global freshwater shortage is driven by socio-economic development, changing consumption patterns, and systemic inefficiencies, with semi-arid regions like Dodoma, Tanzania, being especially vulnerable. Traditional statistical and regression-based models for predicting water deficits have proven insufficient in capturing the complex, nonlinear interactions among climatic, hydrological, and anthropogenic factors. To address this gap, this study proposes a deep learning-based predictive framework utilising advanced algorithms, Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Deep Neural Networks (DNN) to improve the forecasting of water deficits. Using a thirteen-year (13) dataset collected from the semi-arid climate region of Dodoma, encompassing meteorological, hydrological, and socioeconomic variables. The models were trained and evaluated using performance metrics such as Root Mean Square Error (RMSE) and R-squared (R²). The DNN model demonstrated superior performance with an RMSE of 0.049 and an R² of 1.000, significantly outperforming other models. LSTM, CNN, and RNN models showed moderate to weak predictive accuracy, particularly in handling long-term dependencies and extreme deficit events. The key finding of this study is that the DNN model provides highly reliable and accurate water deficit predictions, making it the most effective among the tested deep learning approaches. This result highlights the value of incorporating deep learning into water resource planning, especially in data-scarce, semi-arid regions. The study concludes that DNN-based models should be prioritised for operational deployment in early warning systems and decision-making platforms. Future work should explore hybrid architectures, hyperparameter tuning, and integration with real-time data sources to enhance robustness and applicability

The Internet of Medical Things (IoMT) is an extended version of the Internet of Things (IoT). It mainly concentrates on the integration of medical things for servicing needy people who cannot get medical services easily, especially rural area people and aged peoples living alone. The main objective of this work is to design a real time interactive system for providing medical services to the needy who do not have a sufficient medical infrastructure. With the help of this system, people will get medical services at their end with minimal medical infrastructure and less treatment cost. However, the designed system could be upgraded to address the family of SARs viruses, and for experimentation, we have taken COVID-19 as a test case. The proposed system comprises of many modules, such as the user interface, analytics, cloud, etc. The proposed user interface is designed for interactive data collection. At the initial stage, it collects preliminary medical information, such as the pulse oxygen rate and RT-PCR results. With the help of a pulse oximeter, they could get the pulse oxygen level. With the help of swap test kit, they could find COVID-19 positivity. That information is uploaded as preliminary information to the designed proposed system via the designed UI. If the system identifies the COVID positivity, it requests that the person upload X-ray/CT images for ranking the severity of the disease. The system is designed for multi-model data. Hence, it can deal with X-ray, CT images, and textual data (RT-PCR results). Once X-ray/CT images are collected via the designed UI, those images are forwarded to the designed AI module for analytics. The proposed AI system is designed for multi-disease classification. It classifies the patients affected with COVID-19 or pneumonia or any other viral infection. It also measures the intensity level of lung infection for providing suitable treatment to the patients. Numerous deep convolution neural network (DCNN) architectures are available for medical image classification. We used ResNet-50, ResNet-100, ResNet-101, VGG 16, and VGG 19 for better classification. From the experimentation, it observed that ResNet101 and VGG 19 outperform, with an accuracy of 97% for CT images. ResNet101 outperforms with an accuracy of 98% for X-ray images. For obtaining enhanced accuracy, we used a major voting classifier. It combines all the classifiers result and presents the majority voted one. It results in reduced classifier bias. Finally, the proposed system presents an automatic test summary report textually. It can be accessed via user-friendly graphical user interface (GUI). It results in a reduced report generation time and individual bias.