Human Radiologists Research Articles

Artificial Intelligence in Head and Neck Cancer Diagnosis: A Comprehensive Review with Emphasis on Radiomics, Histopathological, and Molecular Applications

The present review discusses the transformative role of AI in the diagnosis and management of head and neck cancers (HNCs). Methods: It explores how AI technologies, including ML, DL, and CNNs, are applied in various diagnostic tasks, such as medical imaging, molecular profiling, and predictive modeling. Results: This review highlights AI’s ability to improve diagnostic accuracy and efficiency, particularly in analyzing medical images like CT, MRI, and PET scans, where AI sometimes outperforms human radiologists. This paper also emphasizes AI’s application in histopathology, where algorithms assist in whole-slide image (WSI) analysis, tumor-infiltrating lymphocytes (TILs) quantification, and tumor segmentation. AI shows promise in identifying subtle or rare histopathological patterns and enhancing the precision of tumor grading and treatment planning. Furthermore, the integration of AI with molecular and genomic data aids in mutation analysis, prognosis, and personalized treatment strategies. Conclusions: Despite these advancements, the review identifies challenges in AI adoption, such as data standardization and model interpretability, and calls for further research to fully integrate AI into clinical practice for improved patient outcomes.

An Efficient Ensemble Approach for Brain Tumors Classification Using Magnetic Resonance Imaging

Tumors in the brain can be life-threatening, making early and precise detection crucial for effective treatment and improved patient outcomes. Deep learning (DL) techniques have shown significant potential in automating the early diagnosis of brain tumors by analyzing magnetic resonance imaging (MRI), offering a more efficient and accurate approach to classification. Deep convolutional neural networks (DCNNs), which are a sub-field of DL, have the potential to analyze rapidly and accurately MRI data and, as such, assist human radiologists, facilitating quicker diagnoses and earlier treatment initiation. This study presents an ensemble of three high-performing DCNN models, i.e., DenseNet169, EfficientNetB0, and ResNet50, for accurate classification of brain tumors and non-tumor MRI samples. Our proposed ensemble model demonstrates significant improvements over various evaluation parameters compared to individual state-of-the-art (SOTA) DCNN models. We implemented ten SOTA DCNN models, i.e., EfficientNetB0, ResNet50, DenseNet169, DenseNet121, SqueezeNet, ResNet34, ResNet18, VGG16, VGG19, and LeNet5, and provided a detailed performance comparison. We evaluated these models using two learning rates (LRs) of 0.001 and 0.0001 and two batch sizes (BSs) of 64 and 128 and identified the optimal hyperparameters for each model. Our findings indicate that the ensemble approach outperforms individual models, having 92% accuracy, 90% precision, 92% recall, and an F1 score of 91% at a 64 BS and 0.0001 LR. This study not only highlights the superior performance of the ensemble technique but also offers a comprehensive comparison with the latest research.

Future implications of artificial intelligence in lung cancer screening: a systematic review.

The aim of this study was to systematically review the literature to assess the application of AI-based interventions in lung cancer screening, and its future implications. Relevant published literature was screened using PRISMA guidelines across three databases: PubMed, Scopus, and Web of Science. Search terms for article selection included "artificial intelligence," "radiology," "lung cancer," "screening," and "diagnostic." Included studies evaluated the use of AI in lung cancer screening and diagnosis. Twelve studies met the inclusion criteria. All studies concerned the role of AI in lung cancer screening and diagnosis. The AIs demonstrated promising ability across four domains: (1) detection, (2) characterization and differentiation, (3) augmentation of the work of human radiologists, (4) AI implementation of the LUNG-RADS framework and its ability to augment this framework. All studies reported positive results, demonstrating in some cases AI's ability to perform these tasks to a level close to that of human radiologists. The AI systems included in this review were found to be effective screening tools for lung cancer. These findings hold important implications for the future use of AI in lung cancer screening programmes as they may see use as an adjunctive tool for lung cancer screening that would aid in making early and accurate diagnosis. AI-based systems appear to be powerful tools that can assist radiologists with lung cancer screening and diagnosis.

Effectiveness of machine learning and artificial intelligence in formulating radiological reports for faster and more reliable diagnostics

This article explores the transformative impact of machine learning (ML) and artificial intelligence (AI) on radiological investigations. By enhancing diagnostic accuracy and efficiency, AI algorithms especially deep learning models can detect abnormalities in medical images with remarkable precision, often surpassing human radiologists. The integration of AI expedites image analysis and report generation, alleviating radiologist workloads and reducing burnout. AI systems also offer consistent reporting and continuous learning, improving their performance over time through exposure to vast datasets. Despite the significant benefits, challenges such as data privacy, security, and the need for rigorous validation and oversight remain. As AI continues to advance, it is poised to play a crucial role in enhancing patient care and outcomes in radiology.

Deploying artificial intelligence in the detection of adult appendicular and pelvic fractures in the Singapore emergency department after hours: efficacy, cost savings and non-monetary benefits.

Radiology plays an integral role in fracture detection in the emergency department (ED). After hours, when there are fewer reporting radiologists, most radiographs are interpreted by ED physicians. A minority of these interpretations may miss diagnoses, which later require the callback of patients for further management. Artificial intelligence (AI) has been viewed as a potential solution to augment the shortage of radiologists after hours. We explored the efficacy of an AI solution in the detection of appendicular and pelvic fractures for adult radiographs performed after hours at a general hospital ED in Singapore, and estimated the potential monetary and non-monetary benefits. One hundred and fifty anonymised abnormal radiographs were retrospectively collected and fed through an AI fracture detection solution. The radiographs were re-read by two radiologist reviewers and their consensus was established as the reference standard. Cases were stratified based on the concordance between the AI solution and the reviewers' findings. Discordant cases were further analysed based on the nature of the discrepancy into overcall and undercall subgroups. Statistical analysis was performed to evaluate the accuracy, sensitivity and inter-rater reliability of the AI solution. Ninety-two examinations were included in the final study radiograph set. The AI solution had a sensitivity of 98.9%, an accuracy of 85.9% and an almost perfect agreement with the reference standard. An AI fracture detection solution has similar sensitivity to human radiologists in the detection of fractures on ED appendicular and pelvic radiographs. Its implementation offers significant potential measurable cost, manpower and time savings.

Artificial intelligence for diagnosis and prognosis prediction of natural killer/T cell lymphoma using magnetic resonance imaging

Accurate diagnosis and prognosis prediction are conducive to early intervention and improvement of medical care for natural killer/T cell lymphoma (NKTCL). Artificial intelligence (AI)-based systems are developed based on nasopharynx magnetic resonance imaging. The diagnostic systems achieve areas under the curve of 0.905-0.960 in detecting malignant nasopharyngeal lesions and distinguishing NKTCL from nasopharyngeal carcinoma in independent validation datasets. In comparison to human radiologists, the diagnostic systems show higher accuracies than resident radiologists and comparable ones to senior radiologists. The prognostic system shows promising performance in predicting survival outcomes of NKTCL and outperforms several clinical models. For patients with early-stage NKTCL, only the high-risk group benefits from early radiotherapy (hazard ratio= 0.414 vs. late radiotherapy; 95% confidence interval, 0.190-0.900, p= 0.022), while progression-free survival does not differ in the low-risk group. In conclusion, AI-based systems show potential in assisting accurate diagnosis and prognosis prediction and may contribute to therapeutic optimization for NKTCL.

ChatGPT may free time needed by the interventional radiologist for administration / documentation

Motive: Documentation and administration, unpleasant necessities, take a substantial part of the working time in the subspecialty of interventional radiology. With increasing future demand for clinical radiology predicted, time savings from use of text drafting technologies could be a valuable contribution towards our field. Method: Three cases of peripherally inserted central catheter (PICC) line insertion were defined for the present study. The current version of ChatGPT was tasked with drafting reports, following the Radiological Society of North America (RSNA) template. Key results: Score card evaluation by human radiologists indicates that time savings in documentation / administration can be expected without loss of quality from using ChatGPT. Further, automatically generated texts were not assessed to be clearly identifiable as AI-produced. Conclusions: Patients, doctors, and hospital administrators would welcome a reduction of the time that interventional radiologists need for documentation and administration these days. If AI-tools as tested in the present study are brought into clinical application, questions about trust into those systems eg with regard to medical complications will have to be addressed.

RETRACTED ARTICLE: Diagnostic power of ChatGPT 4 in distal radius fracture detection through wrist radiographs

Distal radius fractures rank among the most prevalent fractures in humans, necessitating accurate radiological imaging and interpretation for optimal diagnosis and treatment. In addition to human radiologists, artificial intelligence systems are increasingly employed for radiological assessments. Since 2023, ChatGPT 4 has offered image analysis capabilities, which can also be used for the analysis of wrist radiographs. This study evaluates the diagnostic power of ChatGPT 4 in identifying distal radius fractures, comparing it with a board-certified radiologist, a hand surgery resident, a medical student, and the well-established AI Gleamer BoneView™. Results demonstrate ChatGPT 4’s good diagnostic accuracy (sensitivity 0.88, specificity 0.98, diagnostic power (AUC) 0.93), surpassing the medical student (sensitivity 0.98, specificity 0.72, diagnostic power (AUC) 0.85; p = 0.04) significantly. Nevertheless, the diagnostic power of ChatGPT 4 lags behind the hand surgery resident (sensitivity 0.99, specificity 0.98, diagnostic power (AUC) 0.985; p = 0.014) and Gleamer BoneView™(sensitivity 1.00, specificity 0.98, diagnostic power (AUC) 0.99; p = 0.006). This study highlights the utility and potential applications of artificial intelligence in modern medicine, emphasizing ChatGPT 4 as a valuable tool for enhancing diagnostic capabilities in the field of medical imaging.

Comparison of the Discrimination Performance of AI Scoring and the Brixia Score in Predicting COVID-19 Severity on Chest X-Ray Imaging: Diagnostic Accuracy Study.

The artificial intelligence (AI) analysis of chest x-rays can increase the precision of binary COVID-19 diagnosis. However, it is unknown if AI-based chest x-rays can predict who will develop severe COVID-19, especially in low- and middle-income countries. The study aims to compare the performance of human radiologist Brixia scores versus 2 AI scoring systems in predicting the severity of COVID-19 pneumonia. We performed a cross-sectional study of 300 patients suspected with and with confirmed COVID-19 infection in Jakarta, Indonesia. A total of 2 AI scores were generated using CAD4COVID x-ray software. The AI probability score had slightly lower discrimination (area under the curve [AUC] 0.787, 95% CI 0.722-0.852). The AI score for the affected lung area (AUC 0.857, 95% CI 0.809-0.905) was almost as good as the human Brixia score (AUC 0.863, 95% CI 0.818-0.908). The AI score for the affected lung area and the human radiologist Brixia score had similar and good discrimination performance in predicting COVID-19 severity. Our study demonstrated that using AI-based diagnostic tools is possible, even in low-resource settings. However, before it is widely adopted in daily practice, more studies with a larger scale and that are prospective in nature are needed to confirm our findings.

Auto-segmentation of Adult-Type Diffuse Gliomas: Comparison of Transfer Learning-Based Convolutional Neural Network Model vs. Radiologists

Segmentation of glioma is crucial for quantitative brain tumor assessment, to guide therapeutic research and clinical management, but very time-consuming. Fully automated tools for the segmentation of multi-sequence MRI are needed. We developed and pretrained a deep learning (DL) model using publicly available datasets A (n = 210) and B (n = 369) containing FLAIR, T2WI, and contrast-enhanced (CE)-T1WI. This was then fine-tuned with our institutional dataset (n = 197) containing ADC, T2WI, and CE-T1WI, manually annotated by radiologists, and split into training (n = 100) and testing (n = 97) sets. The Dice similarity coefficient (DSC) was used to compare model outputs and manual labels. A third independent radiologist assessed segmentation quality on a semi-quantitative 5-scale score. Differences in DSC between new and recurrent gliomas, and between uni or multifocal gliomas were analyzed using the Mann–Whitney test. Semi-quantitative analyses were compared using the chi-square test. We found that there was good agreement between segmentations from the fine-tuned DL model and ground truth manual segmentations (median DSC: 0.729, std-dev: 0.134). DSC was higher for newly diagnosed (0.807) than recurrent (0.698) (p < 0.001), and higher for unifocal (0.747) than multi-focal (0.613) cases (p = 0.001). Semi-quantitative scores of DL and manual segmentation were not significantly different (mean: 3.567 vs. 3.639; 93.8% vs. 97.9% scoring ≥ 3, p = 0.107). In conclusion, the proposed transfer learning DL performed similarly to human radiologists in glioma segmentation on both structural and ADC sequences. Further improvement in segmenting challenging postoperative and multifocal glioma cases is needed.

A Dual-Branch Cross-Modality-Attention Network for Thyroid Nodule Diagnosis Based on Ultrasound Images and Contrast-Enhanced Ultrasound Videos.

Contrast-enhanced ultrasound (CEUS) has been extensively employed as an imaging modality in thyroid nodule diagnosis due to its capacity to visualise the distribution and circulation of micro-vessels in organs and lesions in a non-invasive manner. However, current CEUS-based thyroid nodule diagnosis methods suffered from: 1) the blurred spatial boundaries between nodules and other anatomies in CEUS videos, and 2) the insufficient representations of the local structural information of nodule tissues by the features extracted only from CEUS videos. In this paper, we propose a novel dual-branch network with a cross-modality-attention mechanism for thyroid nodule diagnosis by integrating the information from tow related modalities, i.e., CEUS videos and ultrasound image. The mechanism has two parts: US-attention-from-CEUS transformer (UAC-T) and CEUS-attention-from-US transformer (CAU-T). As such, this network imitates the manner of human radiologists by decomposing the diagnosis into two correlated tasks: 1) the spatio-temporal features extracted from CEUS are hierarchically embedded into the spatial features extracted from US with UAC-T for the nodule segmentation; 2) the US spatial features are used to guide the extraction of the CEUS spatio-temporal features with CAU-T for the nodule classification. The two tasks are intertwined in the dual-branch end-to-end network and optimized with the multi-task learning (MTL) strategy. The proposed method is evaluated on our collected thyroid US-CEUS dataset. Experimental results show that our method achieves the classification accuracy of 86.92%, specificity of 66.41%, and sensitivity of 97.01%, outperforming the state-of-the-art methods. As a general contribution in the field of multi-modality diagnosis of diseases, the proposed method has provided an effective way to combine static information with its related dynamic information, improving the quality of deep learning based diagnosis with an additional benefit of explainability.

Intelligent Breast Mass Classification Approach Using Archimedes Optimization Algorithm with Deep Learning on Digital Mammograms.

Breast cancer (BC) has affected many women around the world. To accomplish the classification and detection of BC, several computer-aided diagnosis (CAD) systems have been introduced for the analysis of mammogram images. This is because analysis by the human radiologist is a complex and time-consuming task. Although CAD systems are used to primarily analyze the disease and offer the best therapy, it is still essential to enhance present CAD systems by integrating novel approaches and technologies in order to provide explicit performances. Presently, deep learning (DL) systems are outperforming promising outcomes in the early detection of BC by creating CAD systems executing convolutional neural networks (CNNs). This article presents an Intelligent Breast Mass Classification Approach using the Archimedes Optimization Algorithm with Deep Learning (BMCA-AOADL) technique on Digital Mammograms. The major aim of the BMCA-AOADL technique is to exploit the DL model with a bio-inspired algorithm for breast mass classification. In the BMCA-AOADL approach, median filtering (MF)-based noise removal and U-Net segmentation take place as a pre-processing step. For feature extraction, the BMCA-AOADL technique utilizes the SqueezeNet model with AOA as a hyperparameter tuning approach. To detect and classify the breast mass, the BMCA-AOADL technique applies a deep belief network (DBN) approach. The simulation value of the BMCA-AOADL system has been studied on the MIAS dataset from the Kaggle repository. The experimental values showcase the significant outcomes of the BMCA-AOADL technique compared to other DL algorithms with a maximum accuracy of 96.48%.

Comparison of Radiologists and Deep Learning for US Grading of Hepatic Steatosis.

Background Screening for nonalcoholic fatty liver disease (NAFLD) is suboptimal due to the subjective interpretation of US images. Purpose To evaluate the agreement and diagnostic performance of radiologists and a deep learning model in grading hepatic steatosis in NAFLD at US, with biopsy as the reference standard. Materials and Methods This retrospective study included patients with NAFLD and control patients without hepatic steatosis who underwent abdominal US and contemporaneous liver biopsy from September 2010 to October 2019. Six readers visually graded steatosis on US images twice, 2 weeks apart. Reader agreement was assessed with use of κ statistics. Three deep learning techniques applied to B-mode US images were used to classify dichotomized steatosis grades. Classification performance of human radiologists and the deep learning model for dichotomized steatosis grades (S0, S1, S2, and S3) was assessed with area under the receiver operating characteristic curve (AUC) on a separate test set. Results The study included 199 patients (mean age, 53 years ± 13 [SD]; 101 men). On the test set (n = 52), radiologists had fair interreader agreement (0.34 [95% CI: 0.31, 0.37]) for classifying steatosis grades S0 versus S1 or higher, while AUCs were between 0.49 and 0.84 for radiologists and 0.85 (95% CI: 0.83, 0.87) for the deep learning model. For S0 or S1 versus S2 or S3, radiologists had fair interreader agreement (0.30 [95% CI: 0.27, 0.33]), while AUCs were between 0.57 and 0.76 for radiologists and 0.73 (95% CI: 0.71, 0.75) for the deep learning model. For S2 or lower versus S3, radiologists had fair interreader agreement (0.37 [95% CI: 0.33, 0.40]), while AUCs were between 0.52 and 0.81 for radiologists and 0.67 (95% CI: 0.64, 0.69) for the deep learning model. Conclusion Deep learning approaches applied to B-mode US images provided comparable performance with human readers for detection and grading of hepatic steatosis. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Tuthill in this issue.

Automated diagnosis of pancreatic mucinous and serous cystic neoplasms with modality-fusion deep neural network using multi-modality MRIs.

Pancreatic cystic neoplasms are increasingly diagnosed with the development of medical imaging technology and people's self-care awareness. However, two of their sub-types, serous cystic neoplasms (SCN) and mucinous cystic neoplasms (MCN), are often misclassified from each other. Because SCN is primarily benign and MCN has a high rate of malignant transformation. Distinguishing SCN and MCN is challenging and essential. MRIs have many different modalities, complete with SCN and MCN diagnosis information. With the help of an artificial intelligence-based algorithm, we aimed to propose a multi-modal hybrid deep learning network that can efficiently diagnose SCN and MCN using multi-modality MRIs. A cross-modal feature fusion structure was innovatively designed, combining features of seven modalities to realize the classification of SCN and MCN. 69 Patients with multi-modalities of MRIs were included, and experiments showed performances of every modality. The proposed method with the optimized settings outperformed all other techniques and human radiologists with high accuracy of 75.07% and an AUC of 82.77%. Besides, the proposed disentanglement method outperformed other fusion methods, and delayed contrast-enhanced T1-weighted MRIs proved most valuable in diagnosing SCN and MCN. Through the use of a contemporary artificial intelligence algorithm, physicians can attain high performance in the complex challenge of diagnosing SCN and MCN, surpassing human radiologists to a significant degree.

Performance of ChatGPT, human radiologists, and context-aware ChatGPT in identifying AO codes from radiology reports

While radiologists can describe a fracture’s morphology and complexity with ease, the translation into classification systems such as the Arbeitsgemeinschaft Osteosynthesefragen (AO) Fracture and Dislocation Classification Compendium is more challenging. We tested the performance of generic chatbots and chatbots aware of specific knowledge of the AO classification provided by a vector-index and compared it to human readers. In the 100 radiological reports we created based on random AO codes, chatbots provided AO codes significantly faster than humans (mean 3.2 s per case vs. 50 s per case, p < .001) though not reaching human performance (max. chatbot performance of 86% correct full AO codes vs. 95% in human readers). In general, chatbots based on GPT 4 outperformed the ones based on GPT 3.5-Turbo. Further, we found that providing specific knowledge substantially enhances the chatbot’s performance and consistency as the context-aware chatbot based on GPT 4 provided 71% consistent correct full AO codes for the compared to the 2% consistent correct full AO codes for the generic ChatGPT 4. This provides evidence, that refining and providing specific context to ChatGPT will be the next essential step in harnessing its power.

Natural Language Processing of Radiology Reports: Predicting Metastatic Progression from Text Data

The project’s goal is to extract tumour measurement data from oncological radiology reports with equal, or improved accuracy of a human radiologist. The purpose is to streamline and improve the efficiency of cancer diagnosis, accomplishing this through methods in artificial intelligence. In this experiment, a collection of 85,218 colorectal, and lung radiology reports were used. After loading reports into a data frame, and a BioBERT (bidirectional encoder representations from transformers pre-trained on biomedical corpora) model into a virtual environment, a question is set to be answered by the model where the context of the question is each radiology report’s findings section of the specified organ. These inputs are tokenized and embedded into numerical values to map sentences to vectors of real numbers. Vectors are fed into the model, and an answer is output in natural language for human readability. Answers are stored in the data frame in their corresponding row from which they were derived. The model successfully answered questions about measurements of tumours written in free text in reports where tumours were present, and successfully ignored or did not report in cases where tumours were not present, or measurements were unchanged. In cases with multiple tumours, the model reported exclusively first listed measurements. In an updated version of this model, context will be run through sentence-wise to ensure equal attention to the context entirely. This project is evidence that using one question and selecting the findings portion of a radiology report for one organ as context in a question-answering model built using BioBERT is effective, and efficient in collecting measurements from radiology reports. This algorithm can be applied to other areas of medicine, or other fields entirely with a few model alterations. This project is a step forward in improving cancer diagnosis efficiency and improving medical AI.

New technologies in medicine and healthcare: benefits and drawbacks

Advancements in technology have revolutionized the healthcare industry, enabling us to detect, diagnose,and treat medical conditions with greater accuracy and speed than ever before. From electronic health recordsto telemedicine, these technologies are improving patient outcomes, enhancing communication between healthcareprofessionals, and reducing healthcare costs.One of the most exciting technologies in medicine is Artificial Intelligence (AI). AI algorithms can analyzevast amounts of medical data and provide insights into diseases that would be impossible for humansto identify [1]. For example, AI has been used to identify early signs of cardiovascular diseases and to detectbreast cancer in mammograms with greater accuracy than human radiologists [2]. However, some expertshave expressed concern that AI could replace human doctors, leading to a lack of empathy and a loss of thepersonal touch in healthcare.Telemedicine is another technology that is gaining popularity, particularly in rural areas where access tohealthcare can be limited. Telemedicine allows doctors to consult with patients remotely, using video conferencingand other technologies to diagnose and treat medical conditions [3]. This can save patients time andmoney, as well as reduce the burden on healthcare providers. However, telemedicine is not suitable for allmedical conditions, and there are concerns about privacy and security when it comes to transmitting sensitivemedical information over the internet.Electronic Health Records (EHRs) are another technology that is transforming healthcare. EHRs enablehealthcare providers to access a patient’s medical history, medications, and test results quickly and easily.This can improve patient outcomes by reducing errors and providing doctors with a complete picture of apatient’s health. However, there are concerns about data privacy and security when it comes to storing sensitivemedical information electronically. Some solutions have arrived, as blockchain technologies [4].In conclusion, technology has the potential to transform healthcare and improve patient outcomes. Thesetechnologies offer exciting possibilities for the future of medicine. However, it is important to consider theethical, privacy, and security implications of these technologies and to ensure that they are used in a waythat benefits patients without compromising their rights or safety.

Quantitative Evaluation of COVID-19 Pneumonia CT Using AI Analysis-Feasibility and Differentiation from Other Common Pneumonia Forms.

PURPOSE: To implement the technical feasibility of an AI-based software prototype optimized for the detection of COVID-19 pneumonia in CT datasets of the lung and the differentiation between other etiologies of pneumonia. METHODS: This single-center retrospective case-control-study consecutively yielded 144 patients (58 female, mean age 57.72 ± 18.25 y) with CT datasets of the lung. Subgroups including confirmed bacterial (n = 24, 16.6%), viral (n = 52, 36.1%), or fungal (n = 25, 16.6%) pneumonia and (n = 43, 30.7%) patients without detected pneumonia (comparison group) were evaluated using the AI-based Pneumonia Analysis prototype. Scoring (extent, etiology) was compared to reader assessment. RESULTS: The software achieved an optimal sensitivity of 80.8% with a specificity of 50% for the detection of COVID-19; however, the human radiologist achieved optimal sensitivity of 80.8% and a specificity of 97.2%. The mean postprocessing time was 7.61 ± 4.22 min. The use of a contrast agent did not influence the results of the software (p = 0.81). The mean evaluated COVID-19 probability is 0.80 ± 0.36 significantly higher in COVID-19 patients than in patients with fungal pneumonia (p < 0.05) and bacterial pneumonia (p < 0.001). The mean percentage of opacity (PO) and percentage of high opacity (PHO ≥ -200 HU) were significantly higher in COVID-19 patients than in healthy patients. However, the total mean HU in COVID-19 patients was -679.57 ± 112.72, which is significantly higher than in the healthy control group (p < 0.001). CONCLUSION: The detection and quantification of pneumonia beyond the primarily trained COVID-19 datasets is possible and shows comparable results for COVID-19 pneumonia to an experienced reader. The advantages are the fast, automated segmentation and quantification of the pneumonia foci.

Self-Training with Entropy-Based Mixup for Low-Resource Chest X-ray Classification

Deep learning-based medical image analysis technology has been developed to the extent that it shows an accuracy surpassing the ability of a human radiologist in some tasks. However, data labeling on medical images requires human experts and a great deal of time and expense. Moreover, medical image data usually have an imbalanced distribution for each disease. In particular, in multilabel classification, learning with a small number of labeled data causes overfitting problems. The model easily overfits the limited number of labeled data, while it still underfits the large amount of unlabeled data. In this study, we propose a method that combines entropy-based Mixup and self-training to improve the performance of data-imbalanced chest X-ray classification. The proposed method is to apply the Mixup algorithm to limited labeled data to alleviate the data imbalance problem and perform self-training that effectively utilizes the unlabeled data while iterating this process by replacing the teacher model with the student model. Experimental results in an environment with a limited number of labeled data and a large number of unlabeled data showed that the classification performance was improved by combining entropy-based Mixup and self-training.

Artificial intelligence in lung cancer diagnostic imaging: a review of the reporting and conduct of research published 2018-2019.

This study aimed to describe the methodologies used to develop and evaluate models that use artificial intelligence (AI) to analyse lung images in order to detect, segment (outline borders of), or classify pulmonary nodules as benign or malignant. In October 2019, we systematically searched the literature for original studies published between 2018 and 2019 that described prediction models using AI to evaluate human pulmonary nodules on diagnostic chest images. Two evaluators independently extracted information from studies, such as study aims, sample size, AI type, patient characteristics, and performance. We summarised data descriptively. The review included 153 studies: 136 (89%) development-only studies, 12 (8%) development and validation, and 5 (3%) validation-only. CT scans were the most common type of image type used (83%), often acquired from public databases (58%). Eight studies (5%) compared model outputs with biopsy results. 41 studies (26.8%) reported patient characteristics. The models were based on different units of analysis, such as patients, images, nodules, or image slices or patches. The methods used to develop and evaluate prediction models using AI to detect, segment, or classify pulmonary nodules in medical imaging vary, are poorly reported, and therefore difficult to evaluate. Transparent and complete reporting of methods, results and code would fill the gaps in information we observed in the study publications. We reviewed the methodology of AI models detecting nodules on lung images and found that the models were poorly reported and had no description of patient characteristics, with just a few comparing models' outputs with biopsies results. When lung biopsy is not available, lung-RADS could help standardise the comparisons between the human radiologist and the machine. The field of radiology should not give up principles from the diagnostic accuracy studies, such as the choice for the correct ground truth, just because AI is used. Clear and complete reporting of the reference standard used would help radiologists trust in the performance that AI models claim to have. This review presents clear recommendations about the essential methodological aspects of diagnostic models that should be incorporated in studies using AI to help detect or segmentate lung nodules. The manuscript also reinforces the need for more complete and transparent reporting, which can be helped using the recommended reporting guidelines.