Figure 6 - uploaded by Eric A Walker
Content may be subject to copyright.
Perception and detection of abnormalities by AI and humans. Abnormalities may be perceived by AI algorithms not appreciated by humans and vice versa with others being detected by both.
Source publication
Interpretation of increasingly complex imaging studies involves multiple intricate tasks requiring visual evaluation, cognitive processing, and decision-making. At each stage of this process, there are opportunities for error due to human factors including perceptual and ergonomic conditions. Investigation into the root causes of interpretive error...
Context in source publication
Citations
... It is chosen over other imaging techniques like computed tomography (CT) because it is less expensive, more generally available, and has a high measure for pulmonary tuberculosis. Chest X-rays (CXRs) are widely used by experienced physicians [3]. ...
Pulmonary Tuberculosis (PTB) is a highly dangerous illness that can have severe consequences if left untreated. Researchers and physicians are taking more interest in using chest radiography to automate the investigation of pulmonary tuberculosis. For early detection and treatment, computer-aided diagnosis (CAD) systems are being used to automate the diagnosis of TB. Previous literature shows that many deep learning-based approaches (DL) have been introduced for the classification of pulmonary tuberculosis using chest radiographs. In this study, we introduced a modified convolutional neural network (CNN) model comprised of 21 layers with different architectural layers. This study aims to classify images into TB and Normal. The presented methodology is trained and evaluated on the dataset created by combining multiple publicly available datasets. Results of the proposed modified CNN model have been compared with the seven individual pre-trained state-of-art models, including VGG16, VGG19, ResNet50, InceptionV3, Xception, MobileNetV2, and EfficientNetB7. The comparison demonstrates that the modified CNN model outperformed pre-trained models by achieving an accuracy of 0.9081, precision of 0.9317, recall of 0.9323, F1 score of 0.9307, and AUC of 0.9474.
... In clinical, several factors contribute to the likelihood of errors in abdominal trauma CT diagnosis. These include imaging backlog, understaffing, visual fatigue, and overnight shifts, especially when the caseload exceeds the daily workload [12,13]. Moreover, the complexity of the abdomen, with its multitude of organs, poses challenges for comprehensive diagnosis, especially in cases involving multiple injuries or active bleeding. ...
Background
Abdominal computed tomography (CT) scan is a crucial imaging modality for creating cross-sectional images of the abdominal area, particularly in cases of abdominal trauma, which is commonly encountered in traumatic injuries. However, interpreting CT images is a challenge, especially in emergency. Therefore, we developed a novel deep learning algorithm-based detection method for the initial screening of abdominal internal organ injuries.
Methods
We utilized a dataset provided by the Kaggle competition, comprising 3,147 patients, of which 855 were diagnosed with abdominal trauma, accounting for 27.16% of the total patient population. Following image data pre-processing, we employed a 2D semantic segmentation model to segment the images and constructed a 2.5D classification model to assess the probability of injury for each organ. Subsequently, we evaluated the algorithm’s performance using 5k-fold cross-validation.
Results
With particularly noteworthy performance in detecting renal injury on abdominal CT scans, we achieved an acceptable accuracy of 0.932 (with a positive predictive value (PPV) of 0.888, negative predictive value (NPV) of 0.943, sensitivity of 0.887, and specificity of 0.944). Furthermore, the accuracy for liver injury detection was 0.873 (with PPV of 0.789, NPV of 0.895, sensitivity of 0.789, and specificity of 0.895), while for spleen injury, it was 0.771 (with PPV of 0.630, NPV of 0.814, sensitivity of 0.626, and specificity of 0.816).
Conclusions
The deep learning model demonstrated the capability to identify multiple organ injuries simultaneously on CT scans and holds potential for application in preliminary screening and adjunctive diagnosis of trauma cases beyond abdominal injuries.
... The misdiagnosis rate in identifying fractures ranges from 4 to 9%, and delay in the diagnosis and initiation of treatment causes adverse effects on long-term outcomes in these patients [3][4][5]. A plain radiograph's interpretation or diagnosis error might have several causes, from the reader's Extended author information available on the last page of the article experience level to perception errors in reading the radiograph [6][7][8]. Front-line residents and inexperienced junior doctors are more prone to make diagnostic mistakes while treating many trauma patients in an emergency, because misdiagnosis can occur in a busy emergency room under pressure. Peripheral health systems in developing countries lack specialized medical personnel like radiologists. ...
To evaluate the diagnostic accuracy of artificial intelligence-based algorithms in identifying neck of femur fracture on a plain radiograph.
Systematic review and meta-analysis.
PubMed, Web of science, Scopus, IEEE, and the Science direct databases were searched from inception to 30 July 2023.
Eligible article types were descriptive, analytical, or trial studies published in the English language providing data on the utility of artificial intelligence (AI) based algorithms in the detection of the neck of the femur (NOF) fracture on plain X-ray.
The prespecified primary outcome was to calculate the sensitivity, specificity, accuracy, Youden index, and positive and negative likelihood ratios. Two teams of reviewers (each consisting of two members) extracted the data from available information in each study. The risk of bias was assessed using a mix of the CLAIM (the Checklist for AI in Medical Imaging) and QUADAS-2 (A Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies) criteria.
Of the 437 articles retrieved, five were eligible for inclusion, and the pooled sensitivity of AIs in diagnosing the fracture NOF was 85%, with a specificity of 87%. For all studies, the pooled Youden index (YI) was 0.73. The average positive likelihood ratio (PLR) was 19.88, whereas the negative likelihood ratio (NLR) was 0.17. The random effects model showed an overall odds of 1.16 (0.84–1.61) in the forest plot, comparing the AI system with those of human diagnosis. The overall heterogeneity of the studies was marginal (I2 = 51%). The CLAIM criteria for risk of bias assessment had an overall >70% score.
Artificial intelligence (AI)-based algorithms can be used as a diagnostic adjunct, benefiting clinicians by taking less time and effort in neck of the femur (NOF) fracture diagnosis.
PROSPERO CRD42022375449.
... This value proposition follows business objectives that may identify and reduce threats and adverse factors during medical procedures. HC belongs to a high-risk domain since there are uncertain external factors (E4), including physicians' fatigue, distractions, or cognitive biases [73,74]. AI applications can reduce certain risks by enabling precise decision support, detecting misconduct, reducing emergent side effects, and reducing invasiveness. ...
Artificial intelligence (AI) applications pave the way for innovations in the healthcare (HC) industry. However, their adoption in HC organizations is still nascent as organizations often face a fragmented and incomplete picture of how they can capture the value of AI applications on a managerial level. To overcome adoption hurdles, HC organizations would benefit from understanding how they can capture AI applications’ potential.
We conduct a comprehensive systematic literature review and 11 semi-structured expert interviews to identify, systematize, and describe 15 business objectives that translate into six value propositions of AI applications in HC.
Our results demonstrate that AI applications can have several business objectives converging into risk-reduced patient care, advanced patient care, self-management, process acceleration, resource optimization, and knowledge discovery.
We contribute to the literature by extending research on value creation mechanisms of AI to the HC context and guiding HC organizations in evaluating their AI applications or those of the competition on a managerial level, to assess AI investment decisions, and to align their AI application portfolio towards an overarching strategy.
... However, this is prone to subjective evaluation, expert-dependent, and sometimes inefficient process. Subjective discrepancies in radiograph-based illness diagnosis are unavoidable [5,6]. CXR images of TB patients are sometimes confused with other lung abnormalities of similar patterns [7,8]. ...
Tuberculosis (TB) is a chronic infectious lung disease, which caused the death of about 1.5 million people in 2020 alone. Therefore, it is important to detect TB accurately at an early stage to prevent the infection and associated deaths. Chest X-ray (CXR) is the most popularly used method for TB diagnosis. However, it is difficult to identify TB from CXR images in the early stage, which leads to time-consuming and expensive treatments. Moreover, due to the increase of drug-resistant tuberculosis, the disease becomes more challenging in recent years. In this work, a novel deep learning-based framework is proposed to reliably and automatically distinguish TB, non-TB (other lung infections), and healthy patients using a dataset of 40,000 CXR images. Moreover, a stacking machine learning-based diagnosis of drug-resistant TB using 3037 CXR images of TB patients is implemented. The largest drug-resistant TB dataset will be released to develop a machine learning model for drug-resistant TB detection and stratification. Besides, Score-CAM-based visualization technique was used to make the model interpretable to see where the best performing model learns from in classifying the image. The proposed approach shows an accuracy of 93.32% for the classification of TB, non-TB, and healthy patients on the largest dataset while around 87.48% and 79.59% accuracy for binary classification (drug-resistant vs drug-sensitive TB), and three-class classification (multi-drug resistant (MDR), extreme drug-resistant (XDR), and sensitive TB), respectively, which is the best reported result compared to the literature. The proposed solution can make fast and reliable detection of TB and drug-resistant TB from chest X-rays, which can help in reducing disease complications and spread.
... Solutions to these errors include the increasing investment of time and effort in scan and equipment development and routine image quality control, thereby reducing diagnostic errors caused by equipment issues. Apart from that, most perception errors occur when doctors fail to find a meaningful lesion in images (search error), when a lesion is noted for a short time but not given sufficient attention (recognition error), or when doctors attach importance to the lesions but do not provide the correct diagnosis (decision error) [20,21]. Based on our clinical experience and literature published, we categorize perceptual errors into the following causes. ...
... Many non-professional causes can affect the accuracy of reports during the workflow of radiologists, and fatigue is one of the most important causes [20,27]. For instance, fatigue from lack of sleep has been identified as a contributing factor in many severe accidents [28,29]. ...
Diagnostic imaging is an essential and indispensable part of medical diagnosis and treatment, and diagnostic errors or biases are also common in the department of radiology, sometimes even having a severe impact on the diagnosis and treatment of patients. There are various reasons for diagnostic errors or biases in imaging. In this review, we analyze and summarize the causes of diagnostic imaging errors and biases based on practical cases. We propose solutions for dealing with diagnostic imaging errors and reducing their probability, thereby helping radiologists in their clinical practice.
Critical relevance statement Diagnostic errors or bias contribute to most medical errors in the radiology department. Solutions for dealing with diagnostic imaging errors are pivotal for patients.
Key points
• Diagnostic errors or bias contribute to most medical errors in radiology department.
• Solutions for dealing with diagnostic imaging errors are pivotal for patients.
• This review summarizes the causes of diagnostic errors and offers solutions to them.
Graphical Abstract
... Instead of for each diagnostic error a) requesting an expensive second reading from a human-radiologist; b) or, conversely, each time deriving advice from an AI pathology detection system; these mechanisms can be used in certain situations. There are 2 types of diagnostic error of the radiologist: interpretation and perception [35]. With an error in interpretation, the radiologist notices the pathology, but draws the wrong conclusion about it. ...
... Specifically, normal anatomical structures and existing abnormalities noticed in the medical images during the radiologist's review are documented as Findings. Previous studies have revealed the potential for missed abnormalities due to perceptual errors, which could lead to incorrect or delayed diagnoses [6], [7], and adverse repercussions for the patient. This work aims to reduce these perceptual errors by investigating the adaptation of a state-of-the-art VL model architecture to generate Findings from CXR images, and could, in turn, contribute to an automated radiology reporting system. ...
Among all the sub-sections in a typical radiology report, the Clinical Indications, Findings, and Impression often reflect important details about the health status of a patient. The information included in Impression is also often covered in Findings. While Findings and Impression can be deduced by inspecting the image, Clinical Indications often require additional context. The cognitive task of interpreting medical images remains the most critical and often time-consuming step in the radiology workflow. Instead of generating an end-to-end radiology report, in this paper, we focus on generating the Findings from automated interpretation of medical images, specifically chest X-rays (CXRs). Thus, this work focuses on reducing the workload of radiologists who spend most of their time either writing or narrating the Findings. Unlike past research, which addresses radiology report generation as a single-step image captioning task, we have further taken into consideration the complexity of interpreting CXR images and propose a two-step approach: (a) detecting the regions with abnormalities in the image, and (b) generating relevant text for regions with abnormalities by employing a generative large language model (LLM). This two-step approach introduces a layer of interpretability and aligns the framework with the systematic reasoning that radiologists use when reviewing a CXR.
... This is particularly important, as perceptual errors have been shown to account for 60% to 70% of diagnostic errors 3 with satisfaction of search being one of etiologies. 15 Our study suggests that there may be a benefit in utilizing a second pass review, given the strongly positive survey responses. This suggests that application of the second pass review may have a perceived benefit in mitigating some of the perceptual errors that can occur with a systematic search pattern. ...
Purpose:
Diagnostic errors are common in radiology. The gestalt impression of an image refers to the rapid holistic understanding one formulates about an image and may facilitate improved diagnostic accuracy. The ability to generate a gestalt impression is typically acquired over time and is generally not explicitly taught. Our study aims to assess whether perceptual training using second look and minification technique (SLMT) can help image interpreters formulate a holistic understanding of an image and become more accurate at evaluating medical images.
Approach:
Fourteen healthcare trainees voluntarily participated in a perceptual training module, comparing the differences in detection of nodules and other actionable finding (OAF) on chest radiographs before and after perceptual training intervention. The experimental group received SLMT training, and the control group did not.
Results:
Survey results were positive for all items, with the p-values <0.01. There was improvement in the performance in detection of nodules and OAF in both groups. However, this change was statistically significant only for OAFs in the control group (p-value <0.05) but not the experimental group.
Conclusions:
SLMT training was viewed by participants as an extremely helpful educational tool. Survey results indicated that participants felt the SLMT was a beneficial educational intervention. The experimental group's detection of nodules and OAF improved after SLMT, though not statistically significantly so, which may be related to the small sample size or lack of training effect. Perceptual training using SLMT may help as a useful educational technique, help radiologists identify abnormalities, and improve workflow.
... It is essential to have appropriate ergonomic set-up of the reading workstations. Ergonomic set-up of the reading workstations for radiologists mostly focuses on improving posture and having an appropriate amount of adjustable support while improving comfort 9 . Educate ergonomically to do your job is optimizing work performance, decrease burnout, increase comfort, and well-being of radiologist also to decrease strain on the body, which can be brought about by tedious movement, awkward body positions, hard work, vibration, and rest breaks. ...
... Two types of sound should be addressed are noise from computers and discernable conversation 9 . Methods for noise reduction are the use of textile or soft plastic flooring, sound-absorbing wall and ceiling materials, furniture that does not reflect sound (wood instead of steel) and sound-absorbing wall panels. ...
... Standing up and moving around 9 , change position often A sound workplace is a gainful workplace. After 45-60 min of uninterrupted work, radiologist should stand up and take a couple of moments to stroll a few doors down, get a beverage, glance out the window, anything that gets them out of their seat 9 . A very important work-related musculoskeletal disorders prevention factor is movement. ...
Ergonomics is the study of people at work, also is the discipline that is responsible for the design of workplaces, tools, and tasks, so that they coincide with the physiological, anatomical, psychological characteristics, and the capacities of the workers, including Health professionals, who will be involved. Radiology is at the forefront of specialties which have undergone rapid technological revolution. The modernity of the Radiology and Imaging Departments not only implies the equipment with which radiology and imaging studies are carried out and processed, but also the inclusion, in an extremely critical way, of the site where the radiologist generates the interpretation of the images obtained and the corresponding radiological report is originated. Such equipment must have a useful and functional ergonomic design both for the patient and for the radiologist and technician. The objective of the manuscript is to familiarize the actions to materialize an ergonomic environment that generates advantages in productivity and well-being for the radiologist.
