Integration and evaluation of chest X-ray artificial intelligence in clinical practice.

Purpose: To integrate and evaluate an artificial intelligence (AI) system that assists in checking endotracheal tube (ETT) placement on chest x-rays (CXRs) in clinical practice. Approach: In clinical use over 17 months, 214 CXR images were ordered to check ETT placement with AI assistance by intensive care unit (ICU) physicians. The system was built on the SimpleMind Cognitive AI platform and integrated into a clinical workflow. It automatically identified the ETT and checked its placement relative to the trachea and carina. The ETT overlay and misplacement alert messages generated by the AI system were compared with radiology reports as the reference. A survey study was also conducted to evaluate usefulness of the AI system in clinical practice. Results: The alert messages indicating that either the ETT was misplaced or not detected had a positive predictive value of 42% (21/50) and negative predictive value of 98% (161/164) based on the radiology reports. In the survey, radiologist and ICU physician users indicated that they agreed with the AI outputs and that they were useful. Conclusions: The AI system performance in real-world clinical use was comparable to that seen in previous experiments. Based on this and physician survey results, the system can be deployed more widely at our institution, using insights gained from this evaluation to make further algorithm improvements and quality assurance of the AI system.

Translating AI to Clinical Practice: Overcoming Data Shift with Explainability.

To translate artificial intelligence (AI) algorithms into clinical practice requires generalizability of models to real-world data. One of the main obstacles to generalizability is data shift, a data distribution mismatch between model training and real environments. Explainable AI techniques offer tools to detect and mitigate the data shift problem and develop reliable AI for clinical practice. Most medical AI is trained with datasets gathered from limited environments, such as restricted disease populations and center-dependent acquisition conditions. The data shift that commonly exists in the limited training set often causes a significant performance decrease in the deployment environment. To develop a medical application, it is important to detect potential data shift and its impact on clinical translation. During AI training stages, from premodel analysis to in-model and post hoc explanations, explainability can play a key role in detecting model susceptibility to data shift, which is otherwise hidden because the test data have the same biased distribution as the training data. Performance-based model assessments cannot effectively distinguish the model overfitting to training data bias without enriched test sets from external environments. In the absence of such external data, explainability techniques can aid in translating AI to clinical practice as a tool to detect and mitigate potential failures due to data shift. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.

Angular needle tracker and stabilizer for image-guided interventions.

INTRODUCTION: Minimally invasive image-guided interventions have changed the face of procedural medicine. For these procedures, safety and efficacy depend on precise needle placement. Needle targeting devices help improve the accuracy of needle placement, but their use has not seen broad penetration. Some of these devices are costly and require major modifications to the clinical workflow. In this article, we developed a low-cost, disposable, and easy-to-use angulation tracking device, which was based on a redesigned commercial passive needle holder. MATERIAL AND METHODS: The new design provided real-time angulation information for needle tracking. In this design, two potentiometers were used as angulation sensors, and they were connected to two axes of the passive needle holder's arch structure through a 3 D-printed bridge structure. A control unit included an Arduino Pro Mini, a Bluetooth module, and two rechargeable batteries. The angulation was calculated and communicated in real time to a novel developed smartphone app, where real-time angulation information was displayed for guiding the operator to position the needle to the planned angles. RESULTS: The open-air test results showed that the average errors are 1.03° and 1.08° for left-right angulation and head-foot angulation, respectively. The animal cadaver tests revealed that the novel system had an average angular error of 3.2° and a radial distance error of 3.1 mm. CONCLUSIONS: The accuracy was comparable with some commercially available solutions. The novel and low-cost needle tracking device may find a role as part of a real-time precision approach to both planning and implementation of image-guided therapies.

Continuous Learning AI in Radiology: Implementation Principles and Early Applications.

Artificial intelligence (AI) is becoming increasingly present in radiology and health care. This expansion is driven by the principal AI strengths: automation, accuracy, and objectivity. However, as radiology AI matures to become fully integrated into the daily radiology routine, it needs to go beyond replicating static models, toward discovering new knowledge from the data and environments around it. Continuous learning AI presents the next substantial step in this direction and brings a new set of opportunities and challenges. Herein, the authors discuss the main concepts and requirements for implementing continuous AI in radiology and illustrate them with examples from emerging applications.

Prostate Microstructure in Prostate Cancer Using 3-T MRI with Diffusion-Relaxation Correlation Spectrum Imaging: Validation with Whole-Mount Digital Histopathology.

Background Microstructural MRI has the potential to improve diagnosis and characterization of prostate cancer (PCa), but validation with histopathology is lacking. Purpose To validate ex vivo diffusion-relaxation correlation spectrum imaging (DR-CSI) in the characterization of microstructural tissue compartments in prostate specimens from men with PCa by using registered whole-mount digital histopathology (WMHP) as the reference standard. Materials and Methods Men with PCa who underwent 3-T MRI and robotic-assisted radical prostatectomy between June 2018 and January 2019 were prospectively studied. After prostatectomy, the fresh whole prostate specimens were imaged in patient-specific three-dimensionally printed molds by using 3-T MRI with DR-CSI and were then sliced to create coregistered WMHP slides. The DR-CSI spectral signal component fractions (fA, fB, fC) were compared with epithelial, stromal, and luminal area fractions (fepithelium, fstroma, flumen) quantified in PCa and benign tissue regions. A linear mixed-effects model assessed the correlations between (fA, fB, fC) and (fepithelium, fstroma, flumen), and the strength of correlations was evaluated by using Spearman correlation coefficients. Differences between PCa and benign tissues in terms of DR-CSI signal components and microscopic tissue compartments were assessed using two-sided t tests. Results Prostate specimens from nine men (mean age, 65 years ± 7 [standard deviation]) were evaluated; 20 regions from 17 PCas, along with 20 benign tissue regions of interest, were analyzed. Three DR-CSI spectral signal components (spectral peaks) were consistently identified. The fA, fB, and fC were correlated with fepithelium, fstroma, and flumen (all P < .001), with Spearman correlation coefficients of 0.74 (95% confidence interval [CI]: 0.62, 0.83), 0.80 (95% CI: 0.66, 0.89), and 0.67 (95% CI: 0.51, 0.81), respectively. PCa exhibited differences compared with benign tissues in terms of increased fA (PCa vs benign, 0.37 ± 0.05 vs 0.27 ± 0.06; P < .001), decreased fC (PCa vs benign, 0.18 ± 0.06 vs 0.31 ± 0.13; P = .01), increased fepithelium (PCa vs benign, 0.44 ± 0.13 vs 0.26 ± 0.16; P < .001), and decreased flumen (PCa vs benign, 0.14 ± 0.08 vs 0.27 ± 0.18; P = .004). Conclusion Diffusion-relaxation correlation spectrum imaging signal components correlate with microscopic tissue compartments in the prostate and differ between cancer and benign tissue. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Lee and Hectors in this issue.

Radiology's Information Architecture Could Migrate to One Emulating That of Smartphones.

Diagnostic radiology (DxR), having had successful serial co-evolutions with imaging equipment and PACS, is faced with another. With a backdrop termed "globotics transition," it should create an IT and informatics infrastructure capable of integrating artificial intelligence (AI) into current critical communication functions of PACS and incorporating functions currently residing in balkanized products. DxR will face the challenge of adopting sustaining and disruptive AI innovations simultaneously. In this co-evolution, a major selection force for AI will be increasing the flow of information and patients; "increasing" means faster flow over larger areas defined by geography and content. Larger content includes a broader spectrum of imaging and nonimaging information streams that facilitate medical decision making. Evolution to faster flow will gravitate toward a hierarchical IT architecture consisting of many small channels feeding into fewer larger channels, something potentially difficult for current PACS. Smartphone-like architecture optimized for communication and integration could provide a large-channel backbone and many smaller feeding channels for basic functions, as well as those needing to innovate rapidly. New, more flexible architectures stimulate market competition in which DxR could act as an artificial selection force to influence development of faster increased flow in current PACS companies, in disruptors such as consolidated AI companies, or in entirely new entrants like Apple or Google. In this co-evolution, DxR should be able to stimulate design of a modern communication medium that increases the flow of information and decreases the time and energy necessary to absorb it, thereby creating even more indispensable clinical value for itself.

Selective middle cerebral artery occlusion in the rabbit: Technique and characterization with pathologic findings and multimodal MRI.

BACKGROUND: A reliable animal model of ischemic stroke is vital for pre-clinical evaluation of stroke therapies. We describe a reproducible middle cerebral artery (MCA) embolic occlusion in the French Lop rabbit characterized with multimodal MRI and histopathologic tissue analysis. NEW METHOD: Fluoroscopic-guided microcatheter placement was performed in five consecutive subjects with angiographic confirmation of MCA occlusion with autologous clot. Multimodal MRI was obtained prior to occlusion and up to six hours post after which repeat angiography confirmed sustained occlusion. The brain was harvested for histopathologic examination. RESULTS: Angiography confirmed successful MCA catheterization and durable (>6 h) MCA occlusion in all animals. There was increase of ADC volume over time and variable final core volume presumably related to individual variation in collateral flow. FLAIR hyperintensity indicative of cytotoxic edema and parenchymal contrast enhancement reflective of blood brain barrier disruption was observed over time. Tissue staining of the ischemic brain showed edema and structural alterations consistent with infarction. COMPARISON WITH EXISTING METHODS: This study describes a technique of selective catheterization and embolic occlusion of the MCA in the rabbit with MRI characterization of evolution of ischemia in the model. CONCLUSIONS: We demonstrate the feasibility of a rabbit model of embolic MCA occlusion with angiographic documentation. Serial MR imaging demonstrated changes comparable to those observed in human ischemic stroke, confirmed histopathologically.

Deep transfer learning-based prostate cancer classification using 3 Tesla multi-parametric MRI.

PURPOSE: The purpose of the study was to propose a deep transfer learning (DTL)-based model to distinguish indolent from clinically significant prostate cancer (PCa) lesions and to compare the DTL-based model with a deep learning (DL) model without transfer learning and PIRADS v2 score on 3 Tesla multi-parametric MRI (3T mp-MRI) with whole-mount histopathology (WMHP) validation. METHODS: With IRB approval, 140 patients with 3T mp-MRI and WMHP comprised the study cohort. The DTL-based model was trained on 169 lesions in 110 arbitrarily selected patients and tested on the remaining 47 lesions in 30 patients. We compared the DTL-based model with the same DL model architecture trained from scratch and the classification based on PIRADS v2 score with a threshold of 4 using accuracy, sensitivity, specificity, and area under curve (AUC). Bootstrapping with 2000 resamples was performed to estimate the 95% confidence interval (CI) for AUC. RESULTS: After training on 169 lesions in 110 patients, the AUC of discriminating indolent from clinically significant PCa lesions of the DTL-based model, DL model without transfer learning and PIRADS v2 score ≥ 4 were 0.726 (CI [0.575, 0.876]), 0.687 (CI [0.532, 0.843]), and 0.711 (CI [0.575, 0.847]), respectively, in the testing set. The DTL-based model achieved higher AUC compared to the DL model without transfer learning and PIRADS v2 score ≥ 4 in discriminating clinically significant lesions in the testing set. CONCLUSION: The DeLong test indicated that the DTL-based model achieved comparable AUC compared to the classification based on PIRADS v2 score (p = 0.89).

Building a high-resolution T2-weighted MR-based probabilistic model of tumor occurrence in the prostate.

PURPOSE: We present a method for generating a T2 MR-based probabilistic model of tumor occurrence in the prostate to guide the selection of anatomical sites for targeted biopsies and serve as a diagnostic tool to aid radiological evaluation of prostate cancer. MATERIALS AND METHODS: In our study, the prostate and any radiological findings within were segmented retrospectively on 3D T2-weighted MR images of 266 subjects who underwent radical prostatectomy. Subsequent histopathological analysis determined both the ground truth and the Gleason grade of the tumors. A randomly chosen subset of 19 subjects was used to generate a multi-subject-derived prostate template. Subsequently, a cascading registration algorithm involving both affine and non-rigid B-spline transforms was used to register the prostate of every subject to the template. Corresponding transformation of radiological findings yielded a population-based probabilistic model of tumor occurrence. The quality of our probabilistic model building approach was statistically evaluated by measuring the proportion of correct placements of tumors in the prostate template, i.e., the number of tumors that maintained their anatomical location within the prostate after their transformation into the prostate template space. RESULTS: Probabilistic model built with tumors deemed clinically significant demonstrated a heterogeneous distribution of tumors, with higher likelihood of tumor occurrence at the mid-gland anterior transition zone and the base-to-mid-gland posterior peripheral zones. Of 250 MR lesions analyzed, 248 maintained their original anatomical location with respect to the prostate zones after transformation to the prostate. CONCLUSION: We present a robust method for generating a probabilistic model of tumor occurrence in the prostate that could aid clinical decision making, such as selection of anatomical sites for MR-guided prostate biopsies.

Genomic Adequacy from Solid Tumor Core Needle Biopsies of ex Vivo Tissue and in Vivo Lung Masses: Prospective Study.

Purpose To identify the variables and factors that affect the quantity and quality of nucleic acid yields from imaging-guided core needle biopsy. Materials and Methods This study was approved by the institutional review board and compliant with HIPAA. The authors prospectively obtained 232 biopsy specimens from 74 patients (177 ex vivo biopsy samples from surgically resected masses were obtained from 49 patients and 55 in vivo lung biopsy samples from computed tomographic [CT]-guided lung biopsies were obtained from 25 patients) and quantitatively measured DNA and RNA yields with respect to needle gauge, number of needle passes, and percentage of the needle core. RNA quality was also assessed. Significance of correlations among variables was assessed with analysis of variance followed by linear regression. Conditional probabilities were calculated for projected sample yields. Results The total nucleic acid yield increased with an increase in the number of needle passes or a decrease in needle gauge (two-way analysis of variance, P < .0001 for both). However, contrary to calculated differences in volume yields, the effect of needle gauge was markedly greater than the number of passes. For example, the use of an 18-gauge versus a 20-gauge biopsy needle resulted in a 4.8-5.7 times greater yield, whereas a double versus a single pass resulted in a 2.4-2.8 times greater yield for 18- versus 20-gauge needles, respectively. Ninety-eight of 184 samples (53%) had an RNA integrity number of at least 7 (out of a possible score of 10). Conclusion With regard to optimizing nucleic acid yields in CT-guided lung core needle biopsies used for genomic analysis, there should be a preference for using lower gauge needles over higher gauge needles with more passes. ©RSNA, 2016 Online supplemental material is available for this article. An earlier incorrect version of this article appeared online. This article was corrected on October 21, 2016.

Managing Scale and Innovation in Health IT.

Given the high-intensity interaction between radiology and IT, radiology leadership should understand IT's new, somewhat conflicting, dual roles. Managing large-scale and small-scale projects concurrently has become an important challenge for leaders of health IT (HIT). Historical parallels of this challenge can be drawn from transportation and communication systems, in which a large-scale mind-set is needed to build the initial network, whereas a small-scale mind-set is more useful to develop the content that will traverse this network. Innovation and creativity is a cornerstone of content small-scale thinking, and in HIT, that is what is needed to extract the value from it. However, unlike the early historical transportation and communication examples, the time between the development of the infrastructure and the follow-on, value-rich content is shortened greatly because it has become nearly simultaneous in HIT. Weaving the ability to concomitantly manage both large- and small-scale projects into the fabric of the organizational HIT culture will be critical for its success.

Towards use of MRI-guided ultrasound for treating cerebral vasospasm.

Cerebral vasospasm is a major cause of morbidity and mortality in patients with subarachnoid hemorrhage (SAH), causing delayed neurological deficits in as many as one third of cases. Existing therapy targets induction of cerebral vasodilation through use of various drugs and mechanical means, with a range of observed efficacy. Here, we perform a literature review supporting our hypothesis that transcranially delivered ultrasound may have the ability to induce therapeutic cerebral vasodilation and, thus, may one day be used therapeutically in the context of SAH. Prior studies demonstrate that ultrasound can induce vasodilation in both normal and vasoconstricted blood vessels in peripheral tissues, leading to reduced ischemia and cell damage. Among the proposed mechanisms is alteration of several nitric oxide (NO) pathways, where NO is a known vasodilator. While in vivo studies do not point to a specific physical mechanism, results of in vitro studies favor cavitation induction by ultrasound, where the associated shear stresses likely induce NO production. Two papers discussed the effects of ultrasound on the cerebral vasculature. One study applied clinical transcranial Doppler ultrasound to a rodent complete middle cerebral artery occlusion model and found reduced infarct size. A second involved the application of pulsed ultrasound in vitro to murine brain endothelial cells and showed production of a variety of vasodilatory chemicals, including by-products of arachidonic acid metabolism. In sum, nine reviewed studies demonstrated evidence of either cerebrovascular dilation or elaboration of vasodilatory compounds. Of particular interest, all of the reviewed studies used ultrasound capable of transcranial application: pulsed ultrasound, with carrier frequencies ranging between 0.5 and 2.0 MHz, and intensities not substantially above FDA-approved intensity values. We close by discussing potential specific treatment paradigms of SAH and other cerebral ischemic disorders based on MRI-guided transcranial ultrasound.

A data-driven approach for quality assessment of radiologic interpretations.

Given the increasing emphasis on delivering high-quality, cost-efficient healthcare, improved methodologies are needed to measure the accuracy and utility of ordered diagnostic examinations in achieving the appropriate diagnosis. Here, we present a data-driven approach for performing automated quality assessment of radiologic interpretations using other clinical information (e.g., pathology) as a reference standard for individual radiologists, subspecialty sections, imaging modalities, and entire departments. Downstream diagnostic conclusions from the electronic medical record are utilized as "truth" to which upstream diagnoses generated by radiology are compared. The described system automatically extracts and compares patient medical data to characterize concordance between clinical sources. Initial results are presented in the context of breast imaging, matching 18 101 radiologic interpretations with 301 pathology diagnoses and achieving a precision and recall of 84% and 92%, respectively. The presented data-driven method highlights the challenges of integrating multiple data sources and the application of information extraction tools to facilitate healthcare quality improvement.

Topographical Distribution of Epileptogenic Tubers in Patients With Tuberous Sclerosis Complex.

Tuberous sclerosis complex is a multisystem genetic syndrome often affecting the central nervous system. The purpose of the current study was to identify topographical patterns in the distribution specific to epileptogenic (n = 37) and nonepileptogenic (n = 544) tubers throughout the brain for a cohort of 23 tuberous sclerosis complex patients with a history of seizures. Tubers localized to the inferior parietal lobes, middle frontal lobes, middle temporal lobes, or central sulcus regions were associated with a high frequency of epileptogenic tubers. Epileptogenic tubers occurred statistically more frequently within the inferior parietal lobe and within the central sulcus region in children younger than 1 or between 1 and 3 years old, respectively. Results imply seizure activity in tuberous sclerosis complex patients can be associated with the location of cortical tubers.

The risks of innovation in health care.

Innovation in health care creates risks that are unevenly distributed. An evolutionary analogy using species to represent business models helps categorize innovation experiments and their risks. This classification reveals two qualitative categories: early and late diversification experiments. Early diversification has prolific innovations with high risk because they encounter a "decimation" stage, during which most experiments disappear. Participants face high risk. The few decimation survivors can be sustaining or disruptive according to Christensen's criteria. Survivors enter late diversification, during which they again expand, but within a design range limited to variations of the previous surviving designs. Late diversifications carry lower risk. The exception is when disruptive survivors "diversify," which amplifies their disruption. Health care and radiology will experience both early and late diversifications, often simultaneously. Although oversimplifying Christensen's concepts, early diversifications are likely to deliver disruptive innovation, whereas late diversifications tend to produce sustaining innovations. Current health care consolidation is a manifestation of late diversification. Early diversifications will appear outside traditional care models and physical health care sites, as well as with new science such as molecular diagnostics. They warrant attention because decimation survivors will present both disruptive and sustaining opportunities to radiology. Radiology must participate in late diversification by incorporating sustaining innovations to its value chain. Given the likelihood of disruptive survivors, radiology should seriously consider disrupting itself rather than waiting for others to do so. Disruption entails significant modifications of its value chain, hence, its business model, for which lessons may become available from the pharmaceutical industry's current simultaneous experience with early and late diversifications.

The nature of change.

There is little doubt that health care is facing change. The conventional view of change, based on evolution, is that it is slow, gradual, and generally an evenly paced system change. Unfortunately, it is much more uneven, being burstlike, unpredictable, and, in fact, steplike. This pattern is called punctuated equilibrium, which is well illustrated by the metaphorical picture of the Devil's Staircase. These features call for a reassessment of how to cope with change. In addition to detecting change, responding to it and preparing for it require some understanding of the role of experimentation because the evolution algorithm is simple: experimentation, selection, and replication. Experimentation in radiology forms a continuum ranging from modifying traits to developing variants of diagnostic, interventional, and even new integrated services. We often describe experiments by relating their motives (ie, adaptation and innovation), but complex systems see only experiments available for selection. Experiments generating new services and business models are the important ones because they create the "subspecies" of radiology, which offers a robust set of options capable of withstanding new health care selection forces. Experimentation and selection are the prerequisites of replication (i.e., survival). It behooves radiology to combine and concatenate diversified, reactive, and innovative experiments to explore adjacent domains to expand its set of options. Just as in Darwinian evolution, major changes on the health care landscape will be at the specialty, ie, species and subspecies levels, rather than at the individual specialty trait level. Radiology needs a strong set of "subspecies" to succeed in selection to enhance evolution and allow replication.