Will machine learning end the viability of radiology as a thriving medical specialty?

Evaluating the reliability of ChatGPT as a tool for imaging test referral: a comparative study with a clinical decision support system

OBJECTIVES

As the technology continues to evolve and advance, we can expect to see artificial intelligence (AI) being used in increasingly sophisticated ways to make a diagnosis and decisions such as suggesting the most appropriate imaging referrals. We aim to explore whether Chat Generative Pretrained Transformer (ChatGPT) can provide accurate imaging referrals for clinical use that are at least as good as the ESR iGuide.

METHODS

A comparative study was conducted in a tertiary hospital. Data was collected from 97 consecutive cases that were admitted to the emergency department with abdominal complaints. We compared the imaging test referral recommendations suggested by the ESR iGuide and the ChatGPT and analyzed cases of disagreement. In addition, we selected cases where ChatGPT recommended a chest abdominal pelvis (CAP) CT (n = 66), and asked four specialists to grade the appropriateness of the referral.

RESULTS

ChatGPT recommendations were consistent with the recommendations provided by the ESR iGuide. No statistical differences were found between the appropriateness of referrals by age or gender. Using a sub-analysis of CAP cases, a high agreement between ChatGPT and the specialists was found. Cases of disagreement (12.4%) were further analyzed and presented themes of vague recommendations such as "it would be advisable" and "this would help to rule out."

CONCLUSIONS

ChatGPT's ability to guide the selection of appropriate tests may be comparable to some degree with the ESR iGuide. Features such as the clinical, ethical, and regulatory implications are still warranted and need to be addressed prior to clinical implementation. Further studies are needed to confirm these findings.

CLINICAL RELEVANCE STATEMENT

The article explores the potential of using advanced language models, such as ChatGPT, in healthcare as a CDS for selecting appropriate imaging tests. Using ChatGPT can improve the efficiency of the decision-making process KEY POINTS: • ChatGPT recommendations were highly consistent with the recommendations provided by the ESR iGuide. • ChatGPT's ability in guiding the selection of appropriate tests may be comparable to some degree with ESR iGuide's.

Adopting and expanding ethical principles for generative artificial intelligence from military to healthcare

In 2020, the U.S. Department of Defense officially disclosed a set of ethical principles to guide the use of Artificial Intelligence (AI) technologies on future battlefields. Despite stark differences, there are core similarities between the military and medical service. Warriors on battlefields often face life-altering circumstances that require quick decision-making. Medical providers experience similar challenges in a rapidly changing healthcare environment, such as in the emergency department or during surgery treating a life-threatening condition. Generative AI, an emerging technology designed to efficiently generate valuable information, holds great promise. As computing power becomes more accessible and the abundance of health data, such as electronic health records, electrocardiograms, and medical images, increases, it is inevitable that healthcare will be revolutionized by this technology. Recently, generative AI has garnered a lot of attention in the medical research community, leading to debates about its application in the healthcare sector, mainly due to concerns about transparency and related issues. Meanwhile, questions around the potential exacerbation of health disparities due to modeling biases have raised notable ethical concerns regarding the use of this technology in healthcare. However, the ethical principles for generative AI in healthcare have been understudied. As a result, there are no clear solutions to address ethical concerns, and decision-makers often neglect to consider the significance of ethical principles before implementing generative AI in clinical practice. In an attempt to address these issues, we explore ethical principles from the military perspective and propose the "GREAT PLEA" ethical principles, namely Governability, Reliability, Equity, Accountability, Traceability, Privacy, Lawfulness, Empathy, and Eutonomy, for generative AI in healthcare. Furthermore, we introduce a framework for adopting and expanding these ethical principles in a practical way that has been useful in the military and can be applied to healthcare for generative AI, based on contrasting their ethical concerns and risks. Ultimately, we aim to proactively address the ethical dilemmas and challenges posed by the integration of generative AI into healthcare practice.

Machine learning prediction of motor function in chronic stroke patients: a systematic review and meta-analysis

Background

Recent studies have reported that machine learning (ML), with a relatively strong capacity for processing non-linear data and adaptive ability, could improve the accuracy and efficiency of prediction. The article summarizes the published studies on ML models that predict motor function 3-6 months post-stroke.

Methods

A systematic literature search was conducted in PubMed, Embase, Cochorane and Web of Science as of April 3, 2023 for studies on ML prediction of motor function in stroke patients. The quality of the literature was assessed using the Prediction model Risk Of Bias Assessment Tool (PROBAST). A random-effects model was preferred for meta-analysis using R4.2.0 because of the different variables and parameters.

Results

A total of 44 studies were included in this meta-analysis, involving 72,368 patients and 136 models. Models were categorized into subgroups according to the predicted outcome Modified Rankin Scale cut-off value and whether they were constructed based on radiomics. C-statistics, sensitivity, and specificity were calculated. The random-effects model showed that the C-statistics of all models were 0.81 (95% CI: 0.79; 0.83) in the training set and 0.82 (95% CI: 0.80; 0.85) in the validation set. According to different Modified Rankin Scale cut-off values, C-statistics of ML models predicting Modified Rankin Scale>2(used most widely) in stroke patients were 0.81 (95% CI: 0.78; 0.84) in the training set, and 0.84 (95% CI: 0.81; 0.87) in the validation set. C-statistics of radiomics-based ML models in the training set and validation set were 0.81 (95% CI: 0.78; 0.84) and 0.87 (95% CI: 0.83; 0.90), respectively.

Conclusion

ML can be used as an assessment tool for predicting the motor function in patients with 3-6 months of post-stroke. Additionally, the study found that ML models with radiomics as a predictive variable were also demonstrated to have good predictive capabilities. This systematic review provides valuable guidance for the future optimization of ML prediction systems that predict poor motor outcomes in stroke patients.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022335260, identifier: CRD42022335260.

Artificial Intelligence Curriculum Needs Assessment for a Pediatric Radiology Fellowship Program: What, How, and Why?

RATIONALE AND OBJECTIVES

Artificial intelligence (AI) holds enormous potential for improvements in patient care, efficiency, and innovation in pediatric radiology practice. Although there is a pressing need for a radiology-specific training curriculum and formalized AI teaching, few resources are available. The purpose of our study was to perform a needs assessment for the development of an AI curriculum during pediatric radiology training and continuing education.

MATERIALS AND METHODS

A focus group study using a semistructured moderator-guided interview was conducted with radiology trainees' and attending radiologists' perceptions of AI, perceived competence in interpretation of AI literature, and perceived expectations from radiology AI educational programs. The focus group was audio-recorded, transcribed, and thematic analysis was performed.

RESULTS

The focus group was held virtually with seven participants. The following themes we identified: (1) AI knowledge, (2) previous training, (3) learning preferences, (4) AI expectations, and (5) AI concerns. The participants had no previous formal training in AI and variability in perceived needs and interests. Most preferred a case-based approach to teaching AI. They expressed incomplete understanding of AI hindered its clinical applicability and reiterated a need for improved training in the interpretation and application of AI literature in their practice.

CONCLUSION

We found heterogeneity in perspectives about AI; thus, a curriculum must account for the wide range of these interests and needs. Teaching the interpretation of AI research methods, literature critique, and quality control through implementation of specific scenarios could engage a variety of trainees from different backgrounds and interest levels while ensuring a baseline level of competency in AI.

A novel application of neural networks to identify potentially effective combinations of biologic factors for enhancement of bone fusion/repair

INTRODUCTION

The use of biologic adjuvants (orthobiologics) is becoming commonplace in orthopaedic surgery. Among other applications, biologics are often added to enhance fusion rates in spinal surgery and to promote bone healing in complex fracture patterns. Generally, orthopaedic surgeons use only one biomolecular agent (ie allograft with embedded bone morphogenic protein-2) rather than several agents acting in concert. Bone fusion, however, is a highly multifactorial process and it likely could be more effectively enhanced using biologic factors in combination, acting synergistically. We used artificial neural networks, trained via machine learning on experimental data on orthobiologic interventions and their outcomes, to identify combinations of orthobiologic factors that potentially would be more effective than single agents. This use of machine learning applied to orthobiologic interventions is unprecedented.

METHODS

Available data on the outcomes associated with various orthopaedic biologic agents, electrical stimulation, and pulsed ultrasound were curated from the literature and assembled into a form suitable for machine learning. The best among many different types of neural networks was chosen for its ability to generalize over this dataset, and that network was used to make predictions concerning the expected efficacy of 2400 medically feasible combinations of 9 different agents and treatments.

RESULTS

The most effective combinations were high in the bone-morphogenic proteins (BMP) 2 and 7 (BMP2, 15mg; BMP7, 5mg), and in osteogenin (150ug). In some of the most effective combinations, electrical stimulation could substitute for osteogenin. Some other effective combinations also included bone marrow aspirate concentrate. BMP2 and BMP7 appear to have the strongest pairwise linkage of the factors analyzed in this study.

CONCLUSIONS

Artificial neural networks are powerful forms of artificial intelligence that can be applied readily in the orthopaedic domain, but neural network predictions improve along with the amount of data available to train them. This study provides a starting point from which networks trained on future, expanded datasets can be developed. Yet even this initial model makes specific predictions concerning potentially effective combinatorial therapeutics that should be verified experimentally. Furthermore, our analysis provides an avenue for further research into the basic science of bone healing by demonstrating agents that appear to be linked in function.

Stakeholders' perspectives on the future of artificial intelligence in radiology: a scoping review

OBJECTIVES

Artificial intelligence (AI) has the potential to impact clinical practice and healthcare delivery. AI is of particular significance in radiology due to its use in automatic analysis of image characteristics. This scoping review examines stakeholder perspectives on AI use in radiology, the benefits, risks, and challenges to its integration.

METHODS

A search was conducted from 1960 to November 2019 in EMBASE, PubMed/MEDLINE, Web of Science, Cochrane Library, CINAHL, and grey literature. Publications reflecting stakeholder attitudes toward AI were included with no restrictions.

RESULTS

Commentaries (n = 32), surveys (n = 13), presentation abstracts (n = 8), narrative reviews (n = 8), and a social media study (n = 1) were included from 62 eligible publications. These represent the views of radiologists, surgeons, medical students, patients, computer scientists, and the general public. Seven themes were identified (predicted impact, potential replacement, trust in AI, knowledge of AI, education, economic considerations, and medicolegal implications). Stakeholders anticipate a significant impact on radiology, though replacement of radiologists is unlikely in the near future. Knowledge of AI is limited for non-computer scientists and further education is desired. Many expressed the need for collaboration between radiologists and AI specialists to successfully improve patient care.

CONCLUSIONS

Stakeholder views generally suggest that AI can improve the practice of radiology and consider the replacement of radiologists unlikely. Most stakeholders identified the need for education and training on AI, as well as collaborative efforts to improve AI implementation. Further research is needed to gain perspectives from non-Western countries, non-radiologist stakeholders, on economic considerations, and medicolegal implications.

KEY POINTS

Stakeholders generally expressed that AI alone cannot be used to replace radiologists. The scope of practice is expected to shift with AI use affecting areas from image interpretation to patient care. Patients and the general public do not know how to address potential errors made by AI systems while radiologists believe that they should be "in-the-loop" in terms of responsibility. Ethical accountability strategies must be developed across governance levels. Students, residents, and radiologists believe that there is a lack in AI education during medical school and residency. The radiology community should work with IT specialists to ensure that AI technology benefits their work and centres patients.

Non-radiologist perception of the use of artificial intelligence (AI) in diagnostic medical imaging reports

INTRODUCTION

Incorporating artificial intelligence (AI) in diagnostic medical imaging reports has the potential to improve efficiency. Although perception of radiologists, radiographers, medical students and patients on AI use in image reporting has been explored, there is limited literature on non-radiologist clinicians' opinion on this topic.

METHOD

Single-centre online survey targeting non-radiologist medical staff conducted from May to August 2021 at a tertiary referral hospital in Melbourne, Australia. Survey questions revolved around clinicians' level of comfort acting on AI-generated reports with varying levels of radiologist involvement and scan complexity, opinion on medicolegal responsibility for erroneous AI-issued reports and perception of data privacy and security.

RESULTS

Eighty-eight responses were collected, including 47.9% of consultants. Non-radiologist clinicians across all seniorities and specialties felt significantly less comfortable acting on AI-issued reports compared with radiologist-issued reports (mean comfort radiologist 6.44/7, mean comfort AI 3.35/7, P < 0.001) but felt equally comfortable with an AI-hybrid model of care (mean comfort hybrid 6.38/7, P = 0.676). Non-radiologist clinicians believed that medicolegal responsibility with errors in AI-issued reports mostly lay with hospitals or health service providers (65.9%) and radiologists (54.5%). Regarding data privacy and security, non-radiologist clinicians felt significantly less comfortable with AI issuing image reports instead of radiologists (P < 0.001).

CONCLUSION

A hybrid AI-generated radiologist-confirmed method of image reporting may be the ideal way of integrating AI into clinical practice based on the perception of our referring non-radiologist medical colleagues. Formal guidelines on medicolegal responsibility and data privacy should be established prior to utilising AI in the clinical setting.

Machine Learning in Action: Stroke Diagnosis and Outcome Prediction

The application of machine learning has rapidly evolved in medicine over the past decade. In stroke, commercially available machine learning algorithms have already been incorporated into clinical application for rapid diagnosis. The creation and advancement of deep learning techniques have greatly improved clinical utilization of machine learning tools and new algorithms continue to emerge with improved accuracy in stroke diagnosis and outcome prediction. Although imaging-based feature recognition and segmentation have significantly facilitated rapid stroke diagnosis and triaging, stroke prognostication is dependent on a multitude of patient specific as well as clinical factors and hence accurate outcome prediction remains challenging. Despite its vital role in stroke diagnosis and prognostication, it is important to recognize that machine learning output is only as good as the input data and the appropriateness of algorithm applied to any specific data set. Additionally, many studies on machine learning tend to be limited by small sample size and hence concerted efforts to collate data could improve evaluation of future machine learning tools in stroke. In the present state, machine learning technology serves as a helpful and efficient tool for rapid clinical decision making while oversight from clinical experts is still required to address specific aspects not accounted for in an automated algorithm. This article provides an overview of machine learning technology and a tabulated review of pertinent machine learning studies related to stroke diagnosis and outcome prediction.

Machine learning: principles and applications for thoracic surgery

OBJECTIVES

Machine learning (ML) has experienced a revolutionary decade with advances across many disciplines. We seek to understand how recent advances in ML are going to specifically influence the practice of surgery in the future with a particular focus on thoracic surgery.

METHODS

Review of relevant literature in both technical and clinical domains.

RESULTS

ML is a revolutionary technology that promises to change the way that surgery is practiced in the near future. Spurred by an advance in computing power and the volume of data produced in healthcare, ML has shown remarkable ability to master tasks that had once been reserved for physicians. Supervised learning, unsupervised learning and reinforcement learning are all important techniques that can be leveraged to improve care. Five key applications of ML to cardiac surgery include diagnostics, surgical skill assessment, postoperative prognostication, augmenting intraoperative performance and accelerating translational research. Some key limitations of ML include lack of interpretability, low quality and volumes of relevant clinical data, ethical limitations and difficulties with clinical implementation.

CONCLUSIONS

In the future, the practice of cardiac surgery will be greatly augmented by ML technologies, ultimately leading to improved surgical performance and better patient outcomes.

Glioblastoma Surgery Imaging-Reporting and Data System: Standardized Reporting of Tumor Volume, Location, and Resectability Based on Automated Segmentations

Treatment decisions for patients with presumed glioblastoma are based on tumor characteristics available from a preoperative MR scan. Tumor characteristics, including volume, location, and resectability, are often estimated or manually delineated. This process is time consuming and subjective. Hence, comparison across cohorts, trials, or registries are subject to assessment bias. In this study, we propose a standardized Glioblastoma Surgery Imaging Reporting and Data System (GSI-RADS) based on an automated method of tumor segmentation that provides standard reports on tumor features that are potentially relevant for glioblastoma surgery. As clinical validation, we determine the agreement in extracted tumor features between the automated method and the current standard of manual segmentations from routine clinical MR scans before treatment. In an observational consecutive cohort of 1596 adult patients with a first time surgery of a glioblastoma from 13 institutions, we segmented gadolinium-enhanced tumor parts both by a human rater and by an automated algorithm. Tumor features were extracted from segmentations of both methods and compared to assess differences, concordance, and equivalence. The laterality, contralateral infiltration, and the laterality indices were in excellent agreement. The native and normalized tumor volumes had excellent agreement, consistency, and equivalence. Multifocality, but not the number of foci, had good agreement and equivalence. The location profiles of cortical and subcortical structures were in excellent agreement. The expected residual tumor volumes and resectability indices had excellent agreement, consistency, and equivalence. Tumor probability maps were in good agreement. In conclusion, automated segmentations are in excellent agreement with manual segmentations and practically equivalent regarding tumor features that are potentially relevant for neurosurgical purposes. Standard GSI-RADS reports can be generated by open access software.

A deep learning model for detection of cervical spinal cord compression in MRI scans

Magnetic Resonance Imaging (MRI) evidence of spinal cord compression plays a central role in the diagnosis of degenerative cervical myelopathy (DCM). There is growing recognition that deep learning models may assist in addressing the increasing volume of medical imaging data and provide initial interpretation of images gathered in a primary-care setting. We aimed to develop and validate a deep learning model for detection of cervical spinal cord compression in MRI scans. Patients undergoing surgery for DCM as a part of the AO Spine CSM-NA or CSM-I prospective cohort studies were included in our study. Patients were divided into a training/validation or holdout dataset. Images were labelled by two specialist physicians. We trained a deep convolutional neural network using images from the training/validation dataset and assessed model performance on the holdout dataset. The training/validation cohort included 201 patients with 6588 images and the holdout dataset included 88 patients with 2991 images. On the holdout dataset the deep learning model achieved an overall AUC of 0.94, sensitivity of 0.88, specificity of 0.89, and f1-score of 0.82. This model could improve the efficiency and objectivity of the interpretation of cervical spine MRI scans.

Surgical training fit for the future: the need for a change

Postgraduate training in surgical specialties is one of the longest training programmes in the medical field. Most of the surgical training programmes require 5–6 years of postgraduate training to become qualified. This is usually followed by 1–2 years of fellowship training in a subspecialised interest. This has been the case for the last 20–30 years with no significant change. The surgical practice is transforming quickly due to the advances in medical technology. This transformation is not matched in the postgraduate training, there is minimal exposure to the new technological advances in early years of postgraduate training. The current postgraduate training in surgical specialties is not fit for the future. Early exposure to robotic and artificial intelligence technologies is required. To achieve this, a significant transformation of surgical training is necessary, which requires a new vision and involves significant investment. We discuss the need for this transformation in the postgraduate surgical specialties training and analyse the threats and opportunities in relation to this transformation.

Applications of Machine Learning to Imaging of Spinal Disorders: Current Status and Future Directions

STUDY DESIGN

Narrative review.

OBJECTIVES

We aim to describe current progress in the application of artificial intelligence and machine learning technology to provide automated analysis of imaging in patients with spinal disorders.

METHODS

A literature search utilizing the PubMed database was performed. Relevant studies from all the evidence levels have been included.

RESULTS

Within spine surgery, artificial intelligence and machine learning technologies have achieved near-human performance in narrow image classification tasks on specific datasets in spinal degenerative disease, spinal deformity, spine trauma, and spine oncology.

CONCLUSION

Although substantial challenges remain to be overcome it is clear that artificial intelligence and machine learning technology will influence the practice of spine surgery in the future.

Classification algorithm for congenital Zika Syndrome: characterizations, diagnosis and validation

Zika virus was responsible for the microcephaly epidemic in Brazil which began in October 2015 and brought great challenges to the scientific community and health professionals in terms of diagnosis and classification. Due to the difficulties in correctly identifying Zika cases, it is necessary to develop an automatic procedure to classify the probability of a CZS case from the clinical data. This work presents a machine learning algorithm capable of achieving this from structured and unstructured available data. The proposed algorithm reached 83% accuracy with textual information in medical records and image reports and 76% accuracy in classifying data without textual information. Therefore, the proposed algorithm has the potential to classify CZS cases in order to clarify the real effects of this epidemic, as well as to contribute to health surveillance in monitoring possible future epidemics.

[Diagnostics and Therapy 24/7? Artificial Intelligence as a Challenge and Opportunity in Psychiatry and Psychotherapy]

OBJECTIVE

The aim of the article is to enable a fundamental understanding of the potentials and requirements of Artificial Intelligence (AI) for psychiatrists in the present and for the development of future working environments. Psychiatrists will need to understand the function of AI-systems and personalized AI-assistants in therapy systems and as part of their patients' daily life.

METHOD

The article provides an overview of basic categories and fields of application of AI and machine learning in the diagnosis, prevention and therapy of mental disorders.

RESULTS

AI-applications will shape the prevention, diagnosis and treatment as well as the basic etiological understanding of mental disorders. Notably, the treatment of mental disorders is significantly influenced by commercial product development and assistance systems outside the medical system, as the corresponding developments can exploit large data pools with significantly lower restrictions.

CONCLUSION

Psychiatrists should now seize the opportunity to actively shape the implementation of AI-systems as otherwise key competences could be transferred to a primary field outside the medical system to the detriment of the patient and the therapist.

A survey on Machine learning based Medical Assistive systems in Current Oncological Sciences

BACKGROUND

Cancer is one of the life threatening disease which is affecting a large number of population worldwide. The cancer cells multiply inside the body without showing much symptoms on the surface of the skin thereby making it difficult to predict and detect at the onset of disease. Many organizations are working towards automating the process of cancer detection with minimal false detection rates.

INTRODUCTION

The machine learning algorithms serve to be a promising alternative to support health care practitioners to rule out the disease and predict the growth with various imaging and statistical analysis tools. The medical practitioners are utilizing the output of these algorithms to diagnose and design the course of treatment. These algorithms are capable of finding out the risk level of the patient and can reduce the mortality rate concerning to cancer disease.

METHOD

This article presents the existing state of art techniques for identifying cancer affecting human organs based on machine learning models. The supported set of imaging operations are also elaborated for each type of Cancer.

CONCLUSION

The CAD tools are the aid for the diagnostic radiologists for preliminary investigations and detecting the nature of tumor cells.

Reviewing the relationship between machines and radiology: the application of artificial intelligence

Background

The scope and productivity of artificial intelligence applications in health science and medicine, particularly in medical imaging, are rapidly progressing, with relatively recent developments in big data and deep learning and increasingly powerful computer algorithms. Accordingly, there are a number of opportunities and challenges for the radiological community.

Purpose

To provide review on the challenges and barriers experienced in diagnostic radiology on the basis of the key clinical applications of machine learning techniques.

Material and Methods

Studies published in 2010–2019 were selected that report on the efficacy of machine learning models. A single contingency table was selected for each study to report the highest accuracy of radiology professionals and machine learning algorithms, and a meta-analysis of studies was conducted based on contingency tables.

Results

The specificity for all the deep learning models ranged from 39% to 100%, whereas sensitivity ranged from 85% to 100%. The pooled sensitivity and specificity were 89% and 85% for the deep learning algorithms for detecting abnormalities compared to 75% and 91% for radiology experts, respectively. The pooled specificity and sensitivity for comparison between radiology professionals and deep learning algorithms were 91% and 81% for deep learning models and 85% and 73% for radiology professionals (p < 0.000), respectively. The pooled sensitivity detection was 82% for health-care professionals and 83% for deep learning algorithms (p < 0.005).

Conclusion

Radiomic information extracted through machine learning programs form images that may not be discernible through visual examination, thus may improve the prognostic and diagnostic value of data sets.

Unraveling the deep learning gearbox in optical coherence tomography image segmentation towards explainable artificial intelligence

Machine learning has greatly facilitated the analysis of medical data, while the internal operations usually remain intransparent. To better comprehend these opaque procedures, a convolutional neural network for optical coherence tomography image segmentation was enhanced with a Traceable Relevance Explainability (T-REX) technique. The proposed application was based on three components: ground truth generation by multiple graders, calculation of Hamming distances among graders and the machine learning algorithm, as well as a smart data visualization ('neural recording'). An overall average variability of 1.75% between the human graders and the algorithm was found, slightly minor to 2.02% among human graders. The ambiguity in ground truth had noteworthy impact on machine learning results, which could be visualized. The convolutional neural network balanced between graders and allowed for modifiable predictions dependent on the compartment. Using the proposed T-REX setup, machine learning processes could be rendered more transparent and understandable, possibly leading to optimized applications.

Implementation and design of artificial intelligence in abdominal imaging

Artificial intelligence is a technique that holds promise for helping radiologists improve the care of our patients. At the same time, implementation decisions we make now can have a long-lasting effect on patient outcomes. In the following article, we discuss four areas with unique considerations for implementation of AI: bias, trust, risk, and design. In each section, we highlight applications of AI to abdominal imaging and prostate cancer specifically.

Interventional radiology and artificial intelligence in radiology: Is it time to enhance the vision of our medical students?

OBJECTIVES

To assess awareness and knowledge of Interventional Radiology (IR) in a large population of medical students in 2019.

METHODS

An anonymous survey was distributed electronically to 9546 medical students from first to sixth year at three European medical schools. The survey contained 14 questions, including two general questions on diagnostic radiology (DR) and artificial intelligence (AI), and 11 on IR. Responses were analyzed for all students and compared between preclinical (PCs) (first to third year) and clinical phase (Cs) (fourth to sixth year) of medical school. Of 9546 students, 1459 students (15.3%) answered the survey.

RESULTS

On DR questions, 34.8% answered that AI is a threat for radiologists (PCs: 246/725 (33.9%); Cs: 248/734 (36%)) and 91.1% thought that radiology has a future (PCs: 668/725 (92.1%); Cs: 657/734 (89.5%)). On IR questions, 80.8% (1179/1459) students had already heard of IR; 75.7% (1104/1459) stated that their knowledge of IR wasn't as good as the other specialties and 80% would like more lectures on IR. Finally, 24.2% (353/1459) indicated an interest in a career in IR with a majority of women in preclinical phase, but this trend reverses in clinical phase.

CONCLUSIONS

Development of new technology supporting advances in artificial intelligence will likely continue to change the landscape of radiology; however, medical students remain confident in the need for specialty-trained human physicians in the future of radiology as a clinical practice. A large majority of medical students would like more information about IR in their medical curriculum; almost a quarter of students would be interested in a career in IR.

WITHDRAWN: Radiology Extenders: Impact on Throughput and Accuracy for Routine Chest Radiographs

This article has been withdrawn at the request of the author(s) and/or editor. The Publisher apologizes for any inconvenience this may cause. The full Elsevier Policy on Article Withdrawal can be found at https://www.elsevier.com/about/our-business/policies/article-withdrawal.

Assessing and mitigating the effects of class imbalance in machine learning with application to X-ray imaging

PURPOSE

Machine learning (ML) algorithms are well known to exhibit variations in prediction accuracy when provided with imbalanced training sets typically seen in medical imaging (MI) due to the imbalanced ratio of pathological and normal cases. This paper presents a thorough investigation of the effects of class imbalance and methods for mitigating class imbalance in ML algorithms applied to MI.

METHODS

We first selected five classes from the Image Retrieval in Medical Applications (IRMA) dataset, performed multiclass classification using the random forest model (RFM), and then performed binary classification using convolutional neural network (CNN) on a chest X-ray dataset. An imbalanced class was created in the training set by varying the number of images in that class. Methods tested to mitigate class imbalance included oversampling, undersampling, and changing class weights of the RFM. Model performance was assessed by overall classification accuracy, overall F1 score, and specificity, recall, and precision of the imbalanced class.

RESULTS

A close-to-balanced training set resulted in the best model performance, and a large imbalance with overrepresentation was more detrimental to model performance than underrepresentation. Oversampling and undersampling methods were both effective in mitigating class imbalance, and efficacy of oversampling techniques was class specific.

CONCLUSION

This study systematically demonstrates the effect of class imbalance on two public X-ray datasets on RFM and CNN, making these findings widely applicable as a reference. Furthermore, the methods employed here can guide researchers in assessing and addressing the effects of class imbalance, while considering the data-specific characteristics to optimize imbalance mitigating methods.

Positron Emission Tomography for Response Evaluation in Microenvironment-Targeted Anti-Cancer Therapy

Therapeutic response is evaluated using the diameter of tumors and quantitative parameters of 2-[¹⁸F] fluoro-2-deoxy-d-glucose positron emission tomography (FDG-PET). Tumor response to molecular-targeted drugs and immune checkpoint inhibitors is different from conventional chemotherapy in terms of temporal metabolic alteration and morphological change after the therapy. Cancer stem cells, immunologically competent cells, and metabolism of cancer are considered targets of novel therapy. Accumulation of FDG reflects the glucose metabolism of cancer cells as well as immune cells in the tumor microenvironment, which differs among patients according to the individual immune function; however, FDG-PET could evaluate the viability of the tumor as a whole. On the other hand, specific imaging and cell tracking of cancer cell or immunological cell subsets does not elucidate tumor response in a complexed interaction in the tumor microenvironment. Considering tumor heterogeneity and individual variation in therapeutic response, a radiomics approach with quantitative features of multimodal images and deep learning algorithm with reference to pathologic and genetic data has the potential to improve response assessment for emerging cancer therapy.

Feasibility of Automatic Seed Generation Applied to Cardiac MRI Image Analysis

We present a method of using interactive image segmentation algorithms to reduce specific image segmentation problems to the task of finding small sets of pixels identifying the regions of interest. To this end, we empirically show the feasibility of automatically generating seeds for GrowCut, a popular interactive image segmentation algorithm. The principal contribution of our paper is the proposal of a method for automating the seed generation method for the task of whole-heart segmentation of MRI scans, which achieves competitive unsupervised results (0.76 Dice on the MMWHS dataset). Moreover, we show that segmentation performance is robust to seeds with imperfect precision, suggesting that GrowCut-like algorithms can be applied to medical imaging tasks with little modeling effort.

Identifying Fatal Head Injuries on Postmortem Computed Tomography Using Convolutional Neural Network/Deep Learning: A Feasibility Study

Postmortem computed tomography (PMCT) is a relatively recent advancement in forensic pathology practice that has been increasingly used as an ancillary investigation and screening tool. One area of clinical CT imaging that has garnered a lot of research interest recently is the area of "artificial intelligence" (AI), such as in screening and computer-assisted diagnostics. This feasibility study investigated the application of convolutional neural network, a form of deep learning AI, to PMCT head imaging in differentiating fatal head injury from controls. PMCT images of a transverse section of the head at the level of the frontal sinus from 25 cases of fatal head injury were combined with 25 nonhead-injury controls and divided into training and testing datasets. A convolutional neural network was constructed using Keras and was trained against the training data before being assessed against the testing dataset. The results of this study demonstrated an accuracy of between 70% and 92.5%, with difficulties in recognizing subarachnoid hemorrhage and in distinguishing congested vessels and prominent falx from head injury. These results are promising for potential applications as a screening tool or in computer-assisted diagnostics in the future.

Fetal Weight Estimation Using Automated Fractional Limb Volume With 2-Dimensional Size Parameters: A Multicenter Study

OBJECTIVES

To develop new fetal weight prediction models using automated fractional limb volume (FLV).

METHODS

A prospective multicenter study measured fetal biometry within 4 to 7 days of delivery. Three-dimensional data acquisition included the automated FLV that was based on 50% of the humerus diaphysis (fractional arm volume [AVol]) or 50% of the femur diaphysis (fractional thigh volume [TVol]) length. A regression analysis provided population sample-specific coefficients to develop 4 weight estimation models. Estimated and actual birth weights (BWs) were compared for the mean percent difference ± standard deviation of the percent differences. Systematic errors were analyzed by the Student t test, and random errors were compared by the Pitman test.

RESULTS

A total of 328 pregnancies were scanned before delivery (BW range, 825-5470 g). Only 71.3% to 72.6% of weight estimations were within 10% of actual BW using original published models by Hadlock et al (Am J Obstet Gynecol 1985; 151:333-337) and INTERGROWTH-21st (Ultrasound Obstet Gynecol 2017; 49:478-486). All predictions were accurate by using sample-specific model coefficients to minimize bias in making these comparisons (Hadlock, 0.4% ± 8.7%; INTERGROWTH-21st, 0.5% ± 10.0%; AVol, 0.3% ± 7.4%; and TVol, 0.3% ± 8.0%). Both AVol- and TVol-based models improved the percentage of correctly classified BW ±10% in 83.2% and 83.9% of cases, respectively, compared to the INTERGROWTH-21st model (73.8%; P < .01). For BW of less than 2500 g, all models slightly overestimated BW (+2.0% to +3.1%). For BW of greater than 4000 g, AVol (-2.4% ± 6.5%) and TVol (-2.3% ± 6.9%) models) had weight predictions with small systematic errors that were not different from zero (P > .05). For these larger fetuses, both AVol and TVol models correctly classified BW (±10%) in 83.3% and 87.5% of cases compared to the others (Hadlock, 79.2%; INTERGROWTH-21st, 70.8%) although these differences did not reach statistical significance.

CONCLUSIONS

In this cohort, the inclusion of automated FLV measurements with conventional 2-dimensional biometry was generally associated with improved weight predictions.

Clinical applications and performance of intelligent systems in dental and maxillofacial radiology: A review

Intelligent systems (i.e., artificial intelligence), particularly deep learning, are machines able to mimic the cognitive functions of humans to perform tasks of problem-solving and learning. This field deals with computational models that can think and act intelligently, like the human brain, and construct algorithms that can learn from data to make predictions. Artificial intelligence is becoming important in radiology due to its ability to detect abnormalities in radiographic images that are unnoticed by the naked human eye. These systems have reduced radiologists' workload by rapidly recording and presenting data, and thereby monitoring the treatment response with a reduced risk of cognitive bias. Intelligent systems have an important role to play and could be used by dentists as an adjunct to other imaging modalities in making appropriate diagnoses and treatment plans. In the field of maxillofacial radiology, these systems have shown promise for the interpretation of complex images, accurate localization of landmarks, characterization of bone architecture, estimation of oral cancer risk, and the assessment of metastatic lymph nodes, periapical pathologies, and maxillary sinus pathologies. This review discusses the clinical applications and scope of intelligent systems such as machine learning, artificial intelligence, and deep learning programs in maxillofacial imaging.

Artificial Intelligence and Machine Learning for HIV Prevention: Emerging Approaches to Ending the Epidemic

PURPOSE OF REVIEW

We review applications of artificial intelligence (AI), including machine learning (ML), in the field of HIV prevention.

RECENT FINDINGS

ML approaches have been used to identify potential candidates for preexposure prophylaxis (PrEP) in healthcare settings in the USA and Denmark and in a population-based research setting in Eastern Africa. Although still in the proof-of-concept stage, other applications include ML with smartphone-collected and social media data to promote real-time HIV risk reduction, virtual reality tools to facilitate HIV serodisclosure, and chatbots for HIV education. ML has also been used for causal inference in HIV prevention studies. ML has strong potential to improve delivery of PrEP, with this approach moving from development to implementation. Development and evaluation of AI and ML strategies for HIV prevention may benefit from an implementation science approach, including qualitative assessments with end users, and should be developed and evaluated with attention to equity.

Integrating radiomics into holomics for personalised oncology: from algorithms to bedside

Radiomics, artificial intelligence, and deep learning figure amongst recent buzzwords in current medical imaging research and technological development. Analysis of medical big data in assessment and follow-up of personalised treatments has also become a major research topic in the area of precision medicine. In this review, current research trends in radiomics are analysed, from handcrafted radiomics feature extraction and statistical analysis to deep learning. Radiomics algorithms now include genomics and immunomics data to improve patient stratification and prediction of treatment response. Several applications have already shown conclusive results demonstrating the potential of including other “omics” data to existing imaging features. We also discuss further challenges of data harmonisation and management infrastructure to shed a light on the much-needed integration of radiomics and all other “omics” into clinical workflows. In particular, we point to the emerging paradigm shift in the implementation of big data infrastructures to facilitate databanks growth, data extraction and the development of expert software tools. Secured access, sharing, and integration of all health data, called “holomics”, will accelerate the revolution of personalised medicine and oncology as well as expand the role of imaging specialists.

Artificial Intelligence: reshaping the practice of radiological sciences in the 21st century

Advances in computing hardware and software platforms have led to the recent resurgence in artificial intelligence (AI) touching almost every aspect of our daily lives by its capability for automating complex tasks or providing superior predictive analytics. AI applications are currently spanning many diverse fields from economics to entertainment, to manufacturing, as well as medicine. Since modern AI's inception decades ago, practitioners in radiological sciences have been pioneering its development and implementation in medicine, particularly in areas related to diagnostic imaging and therapy. In this anniversary article, we embark on a journey to reflect on the learned lessons from past AI's chequered history. We further summarize the current status of AI in radiological sciences, highlighting, with examples, its impressive achievements and effect on re-shaping the practice of medical imaging and radiotherapy in the areas of computer-aided detection, diagnosis, prognosis, and decision support. Moving beyond the commercial hype of AI into reality, we discuss the current challenges to overcome, for AI to achieve its promised hope of providing better precision healthcare for each patient while reducing cost burden on their families and the society at large.

Open access image repositories: high-quality data to enable machine learning research

Originally motivated by the need for research reproducibility and data reuse, large-scale, open access information repositories have become key resources for training and testing of advanced machine learning applications in biomedical and clinical research. To be of value, such repositories must provide large, high-quality data sets, where quality is defined as minimising variance due to data collection protocols and data misrepresentations. Curation is the key to quality. We have constructed a large public access image repository, The Cancer Imaging Archive, dedicated to the promotion of open science to advance the global effort to diagnose and treat cancer. Drawing on this experience and our experience in applying machine learning techniques to the analysis of radiology and pathology image data, we will review the requirements placed on such information repositories by state-of-the-art machine learning applications and how these requirements can be met.

Artificial Intelligence: Can Information be Transformed into Intelligence in Surgical Education?

Artificial intelligence (AI) is being rapidly integrated into various medical applications. Although early application of AI has been achieved in image-based, as well as statistical computational models, translation into procedure-based specialties such as surgery may take longer to achieve. A potential application of AI in surgical education is as a teaching coach or mentor that interacts with the used via virtual and/or augmented reality. The question arises as to whether machines will achieve the wisdom and intelligence of human educators.

Finding Inspiration for the Next Generation of Breast Radiologists

Challenges currently facing breast radiologists, including controversial screening mammography guidelines, radiologist burnout, and the perceived threat posed by artificial intelligence could deter potential candidates from pursuing a career in radiology. However, breast radiologists play a fulfilling role by decreasing the effect of breast cancer through both early detection and direct interaction with patients and interdisciplinary clinical colleagues. While perception is that artificial intelligence will threaten the need for radiologists, it is more likely that it will improve image interpretation and efficiency in workflow, thereby further improving patient care. Trainees can be engaged in breast imaging through interactive teaching methods and by role modeling clinical and image interpretation skills.

Physician centred imaging interpretation is dying out - why should I be a nuclear medicine physician?

Radiomics, machine learning, and, more generally, artificial intelligence (AI) provide unique tools to improve the performances of nuclear medicine in all aspects. They may help rationalise the operational organisation of imaging departments, optimise resource allocations, and improve image quality while decreasing radiation exposure and maintaining qualitative accuracy. There is already convincing data that show AI detection, and interpretation algorithms can perform with equal or higher diagnostic accuracy in various specific indications than experts in the field. Preliminary data strongly suggest that AI will be able to process imaging data and information well beyond what is visible to the human eye, and it will be able to integrate features to provide signatures that may further drive personalised medicine. As exciting as these prospects are, they currently remain essentially projects with a long way to go before full validation and routine clinical implementation. AI uses a language that is totally unfamiliar to nuclear medicine physicians, who have not been trained to manage the highly complex concepts that rely primarily on mathematics, computer sciences, and engineering. Nuclear medicine physicians are mostly familiar with biology, pharmacology, and physics, yet, considering the disruptive nature of AI in medicine, we need to start acquiring the knowledge that will keep us in the position of being actors and not merely witnesses of the wonders developed by other stakeholders in front of our incredulous eyes. This will allow us to remain a useful and valid interface between the image, the data, and the patients and free us to pursue other, one might say nobler tasks, such as treating, caring and communicating with our patients or conducting research and development.

Fostering a Healthy AI Ecosystem for Radiology: Conclusions of the 2018 RSNA Summit on AI in Radiology

The 2018 RSNA Summit on AI in Radiology brought together a diverse group of stakeholders to identify and prioritize areas of need related to artificial intelligence in radiology. This article presents the proceedings of the summit with emphasis on RSNA's role in leading, organizing, and catalyzing change during this important time in radiology. © RSNA, 2019.

Artificial Intelligence for the Otolaryngologist: A State of the Art Review

Objective

To provide a state of the art review of artificial intelligence (AI), including its subfields of machine learning and natural language processing, as it applies to otolaryngology and to discuss current applications, future impact, and limitations of these technologies.

Data Sources

PubMed and Medline search engines.

Review Methods

A structured search of the current literature was performed (up to and including September 2018). Search terms related to topics of AI in otolaryngology were identified and queried to identify relevant articles.

Conclusions

AI is at the forefront of conversation in academic research and popular culture. In recent years, it has been touted for its potential to revolutionize health care delivery. Yet, to date, it has made few contributions to actual medical practice or patient care. Future adoption of AI technologies in otolaryngology practice may be hindered by misconceptions of what AI is and a fear that machine errors may compromise patient care. However, with potential clinical and economic benefits, it is vital for otolaryngologists to understand the principles and scope of AI.

Implications for Practice

In the coming years, AI is likely to have a major impact on biomedical research and the practice of medicine. Otolaryngologists are key stakeholders in the development and clinical integration of meaningful AI technologies that will improve patient care. High-quality data collection is essential for the development of AI technologies, and otolaryngologists should seek opportunities to collaborate with data scientists to guide them toward the most impactful clinical questions.