1
|
Xu L, McCandless L, Miller N, Alessio A, Morrison J. Machine-Learned Algorithms to Predict the Risk of Pneumothorax Requiring Chest Tube Placement after Lung Biopsy. J Vasc Interv Radiol 2023; 34:2155-2161. [PMID: 37619941 DOI: 10.1016/j.jvir.2023.08.016] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/23/2022] [Revised: 06/29/2023] [Accepted: 08/15/2023] [Indexed: 08/26/2023] Open
Abstract
PURPOSE To develop a machine-learned algorithm to predict the risk of postlung biopsy pneumothorax requiring chest tube placement (CTP) to facilitate preprocedural decision making, optimize patient care, and improve resource allocation. MATERIALS AND METHODS This retrospective study collected clinical and imaging features of biopsy samples obtained from patients with lung nodule biopsy and included information from 59 procedures resulting in pneumothorax requiring CTP and randomly selected 67 procedures without CTP (convenience sample). The data were divided into 70 and 30 as training and testing sets, respectively. Conventional machine-learned binary classifiers were explored with preprocedural imaging and clinical data as input features and CTP as the output. RESULTS There was no single pathognomonic imaging or predictive clinical feature. For the independent test set under the high-specificity mode, a decision tree, logistic regression, and Naïve Bayes classifier achieved accuracies of identifying CTP at 0.79, 0.93, and 0.89 and area under receiver operating curves (AUROCs) of 0.68, 0.76, and 0.82, respectively. Under high-sensitivity mode, a decision tree, logistic regression, and Naïve Bayes achieved accuracies of identifying CTP of 0.60, 0.45, and 0.60 with AUROCs of 0.71, 0.81, and 0.82, respectively. High importance features included lesion character, chronic obstructive pulmonary disease, lesion depth, and age. A coarse decision tree requiring 4 inputs achieved comparable performance as other methods and previous machine learning prediction studies. CONCLUSIONS The results support the possibility of predicting pneumothorax requiring CTP after biopsy based on an automated decision support, reliant on readily available preprocedural information.
Collapse
Affiliation(s)
- Lu Xu
- Biomedical Engineering, Michigan State University, East Lansing, Michigan; Institute for Quantitative Health Science and Engineering, Michigan State University, East Lansing, Michigan; College of Human Medicine, Michigan State University, East Lansing, Michigan.
| | - Lane McCandless
- College of Human Medicine, Michigan State University, East Lansing, Michigan
| | - Nicholas Miller
- College of Human Medicine, Michigan State University, East Lansing, Michigan
| | - Adam Alessio
- Biomedical Engineering, Michigan State University, East Lansing, Michigan; Institute for Quantitative Health Science and Engineering, Michigan State University, East Lansing, Michigan
| | - James Morrison
- College of Human Medicine, Michigan State University, East Lansing, Michigan; Advanced Radiology Services, Grand Rapids, Michigan
| |
Collapse
|
2
|
Kim DH, Gilyard S, Suh R. Special Considerations and Techniques of Interventions in Lung Transplant Recipients. Tech Vasc Interv Radiol 2023; 26:100926. [PMID: 38123291 DOI: 10.1016/j.tvir.2023.100926] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/23/2023]
Abstract
Lung transplant remains an important treatment option for patients with end-stage lung diseases providing improvement in survival rates and quality of life. Specialized considerations should be applied with interventions of lung transplant recipients as they host specific anatomic variations and high risk towards certain complications. In this article, we highlight the role of interventional radiology for lung transplant recipients along with discussion of interventional techniques. Specific emphasis is placed on describing and explaining the techniques pertained to the points of anastomosis, diagnosis and treatment of malignancies, and management of complications in lung transplant recipients.
Collapse
Affiliation(s)
- Daniel H Kim
- Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA
| | - Shenise Gilyard
- Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA
| | - Robert Suh
- Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA.
| |
Collapse
|
3
|
Kisting MA, Hinshaw JL, Toia GV, Ziemlewicz TJ, Kisting AL, Lee FT, Wagner MG. Artificial Intelligence-Aided Selection of Needle Pathways: Proof-of-Concept in Percutaneous Lung Biopsies. J Vasc Interv Radiol 2023:S1051-0443(23)00830-8. [PMID: 38008378 DOI: 10.1016/j.jvir.2023.11.016] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/03/2023] [Revised: 11/09/2023] [Accepted: 11/17/2023] [Indexed: 11/28/2023] Open
Abstract
PURPOSE To evaluate the concordance between lung biopsy puncture pathways determined by artificial intelligence (AI) and those determined by expert physicians. MATERIALS AND METHODS An AI algorithm was created to choose optimal lung biopsy pathways based on segmented thoracic anatomy and emphysema in volumetric lung computed tomography (CT) scans combined with rules derived from the medical literature. The algorithm was validated using pathways generated from CT scans of randomly selected patients (n = 48) who had received percutaneous lung biopsies and had noncontrast CT scans of 1.25-mm thickness available in picture archiving and communication system (PACS) (n = 28, mean age, 68.4 years ± 9.2; 12 women, 16 men). The algorithm generated 5 potential pathways per scan, including the computer-selected best pathway and 4 random pathways (n = 140). Four experienced physicians rated each pathway on a 1-5 scale, where scores of 1-3 were considered safe and 4-5 were considered unsafe. Concordance between computer and physician ratings was assessed using Cohen's κ. RESULTS The algorithm ratings were statistically equivalent to the physician ratings (safe vs unsafe: κ¯=0.73; ordinal scale: κ¯=0.62). The computer and physician ratings were identical in 57.9% (81/140) of cases and differed by a median of 0 points. All least-cost "best" pathways generated by the algorithm were considered safe by both computer and physicians (28/28) and were judged by physicians to be ideal or near ideal. CONCLUSIONS AI-generated lung biopsy puncture paths were concordant with expert physician reviewers and considered safe. A prospective comparison between computer- and physician-selected puncture paths appears indicated in addition to expansion to other anatomic locations and procedures.
Collapse
Affiliation(s)
- Meridith A Kisting
- Departments of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | - J Louis Hinshaw
- Departments of Radiology, University of Wisconsin-Madison, Madison, Wisconsin; Urology, University of Wisconsin-Madison, Madison, Wisconsin
| | - Giuseppe V Toia
- Departments of Radiology, University of Wisconsin-Madison, Madison, Wisconsin; Medical Physics, University of Wisconsin-Madison, Madison, Wisconsin
| | | | - Adrienne L Kisting
- Departments of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | - Fred T Lee
- Departments of Radiology, University of Wisconsin-Madison, Madison, Wisconsin; Urology, University of Wisconsin-Madison, Madison, Wisconsin; Biomedical Engineering, University of Wisconsin-Madison, Madison, Wisconsin
| | - Martin G Wagner
- Departments of Radiology, University of Wisconsin-Madison, Madison, Wisconsin; Medical Physics, University of Wisconsin-Madison, Madison, Wisconsin.
| |
Collapse
|
4
|
Sorensen AM, Zlevor AM, Kisting MA, Couillard AB, Ziemlewicz TJ, Toia GV, Hinshaw JL, Woods M, Stratchko LM, Pickhardt PJ, Foltz ML, Peppler WW, Lee FT, Knavel Koepsel EM. CT Navigation for Percutaneous Needle Placement: How I Do It. Tech Vasc Interv Radiol 2023; 26:100911. [PMID: 38071032 DOI: 10.1016/j.tvir.2023.100911] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/18/2023]
Abstract
CT navigation (CTN) has recently been developed to combine many of the advantages of conventional CT and CT-fluoroscopic guidance for needle placement. CTN systems display real-time needle position superimposed on a CT dataset. This is accomplished by placing electromagnetic (EM) or optical transmitters/sensors on the patient and needle, combined with fiducials placed within the scan field to superimpose a known needle location onto a CT dataset. Advantages of CTN include real-time needle tracking using a contemporaneous CT dataset with the patient in the treatment position, reduced radiation to the physician, facilitation of procedures outside the gantry plane, fewer helical scans during needle placement, and needle guidance based on diagnostic-quality CT datasets. Limitations include the display of a virtual (vs actual) needle position, which can be inaccurate if the needle bends, the fiducial moves, or patient movement occurs between scans, and limitations in anatomical regions with a high degree of motion such as the lung bases. This review summarizes recently introduced CTN technologies in comparison to historical methods of CT needle guidance. A "How I do it" section follows, which describes how CT navigation has been integrated into the study center for both routine and challenging procedures, and includes step-by-step explanations, technical tips, and pitfalls.
Collapse
Affiliation(s)
- Anna M Sorensen
- Departments of Radiology, University of Wisconsin, Madison, WI
| | - Annie M Zlevor
- Departments of Radiology, University of Wisconsin, Madison, WI
| | | | | | | | - Giuseppe V Toia
- Departments of Radiology, University of Wisconsin, Madison, WI; Medical Physics, University of Wisconsin, Madison, WI
| | - J Louis Hinshaw
- Departments of Radiology, University of Wisconsin, Madison, WI; Departments of Urology, University of Wisconsin, Madison, WI
| | - Michael Woods
- Departments of Radiology, University of Wisconsin, Madison, WI
| | | | | | - Marcia L Foltz
- Departments of Radiology, University of Wisconsin, Madison, WI
| | | | - Fred T Lee
- Departments of Radiology, University of Wisconsin, Madison, WI; Departments of Urology, University of Wisconsin, Madison, WI; Biomedical Engineering, University of Wisconsin, Madison, WI
| | | |
Collapse
|
5
|
Zlevor AM, Kisting MA, Couillard AB, Rossebo AE, Szczykutowicz TP, Mao L, White JK, Hartung MP, Gettle LM, Hinshaw JL, Pickhardt PJ, Ziemlewicz TJ, Foltz ML, Lee FT. Percutaneous CT-Guided Abdominal and Pelvic Biopsies: Comparison of an Electromagnetic Navigation System and CT Fluoroscopy. J Vasc Interv Radiol 2023; 34:910-918. [PMID: 36736821 DOI: 10.1016/j.jvir.2023.01.024] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/06/2022] [Revised: 01/09/2023] [Accepted: 01/22/2023] [Indexed: 02/04/2023] Open
Abstract
PURPOSE To compare electromagnetic navigation (EMN) with computed tomography (CT) fluoroscopy for guiding percutaneous biopsies in the abdomen and pelvis. MATERIALS AND METHODS A retrospective matched-cohort design was used to compare biopsies in the abdomen and pelvis performed with EMN (consecutive cases, n = 50; CT-Navigation; Imactis, Saint-Martin-d'Hères, France) with those performed with CT fluoroscopy (n = 100). Cases were matched 1:2 (EMN:CT fluoroscopy) for target organ and lesion size (±10 mm). RESULTS The population was well-matched (age, 65 vs 65 years; target size, 2.0 vs 2.1 cm; skin-to-target distance, 11.4 vs 10.7 cm; P > .05, EMN vs CT fluoroscopy, respectively). Technical success (98% vs 100%), diagnostic yield (98% vs 95%), adverse events (2% vs 5%), and procedure time (33 minutes vs 31 minutes) were not statistically different (P > .05). Operator radiation dose was less with EMN than with CT fluoroscopy (0.04 vs 1.2 μGy; P < .001), but patient dose was greater (30.1 vs 9.6 mSv; P < .001) owing to more helical scans during EMN guidance (3.9 vs 2.1; P < .001). CT fluoroscopy was performed with a mean of 29.7 tap scans per case. In 3 (3%) cases, CT fluoroscopy was performed with gantry tilt, and the mean angle out of plane for EMN cases was 13.4°. CONCLUSIONS Percutaneous biopsies guided by EMN and CT fluoroscopy were closely matched for technical success, diagnostic yield, procedure time, and adverse events in a matched cohort of patients. EMN cases were more likely to be performed outside of the gantry plane. Radiation dose to the operator was higher with CT fluoroscopy, and patient radiation dose was higher with EMN. Further study with a wider array of procedures and anatomic locations is warranted.
Collapse
Affiliation(s)
- Annie M Zlevor
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | - Meridith A Kisting
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | | | - Annika E Rossebo
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | - Timothy P Szczykutowicz
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin; Department of Medical Physics, University of Wisconsin-Madison, Madison, Wisconsin; Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, Wisconsin
| | - Lu Mao
- Department of Biostatistics and Medical Informatics, University of Wisconsin-Madison, Madison, Wisconsin
| | - James K White
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | - Michael P Hartung
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | | | - J Louis Hinshaw
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin; Department of Urology, University of Wisconsin-Madison, Madison, Wisconsin
| | - Perry J Pickhardt
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | | | - Marcia L Foltz
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin
| | - Fred T Lee
- Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin; Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, Wisconsin; Department of Urology, University of Wisconsin-Madison, Madison, Wisconsin.
| |
Collapse
|
6
|
Wu D, Liu YY, Wang T, Huang YY, Xia P. Computed tomography-guided lung biopsy with rapid on-site evaluation for diagnosis of lung lesions: a meta-analysis. J Cardiothorac Surg 2023; 18:122. [PMID: 37038189 PMCID: PMC10088145 DOI: 10.1186/s13019-023-02212-6] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/01/2022] [Accepted: 04/02/2023] [Indexed: 04/12/2023] Open
Abstract
BACKGROUND Lung biopsy (LB) procedures performed with computed tomography (CT guidance can enable the reliable diagnosis of lung lesions. These diagnostic efforts can be further expedited through a rapid on-site evaluation (ROSE) approach, allowing for the rapid assessment of collected tissue samples to gauge the adequacy of these samples, their features, and associated cytomorphological characteristics. The present analysis was developed to examine the safety and efficacy of CT-guided LB with ROSE as a means of diagnosing lung lesions. METHODS Studies published as of July 31, 2022 in the PubMed, Embase, and Wanfang databases were identified for this meta-analysis. Diagnostic accuracy was the primary endpoint, while secondary endpoints included the operative duration, the number of punctures, and rates of lung hemorrhage, pneumothorax, and secondary LB. RESULTS This meta-analysis included 6 total studies. Relative to CT alone, CT with ROSE was associated with a significant increase in diagnostic accuracy (P < 0.00001). In contrast, there were no significant differences between these two groups with respect to the operative duration (P = 0.86), the number of punctures (P = 0.60), or the rates of pneumothorax (P = 0.82) or lung hemorrhage (P = 0.81). Pooled secondary LB rates were significantly lower for patients that underwent CT with ROSE relative to patients in the CT only group (P = 0.0008). Significant heterogeneity was detected for the operative duration (I2 = 94%) and number of punctures (I2 = 98%) endpoints, while no publication bias was detected for any study endpoints. CONCLUSIONS These results suggest that ROSE may contribute to significant improvements in the diagnostic accuracy of CT-guided LB without contributing to higher rates of complications.
Collapse
Affiliation(s)
- Di Wu
- Department of Pathology, Xuzhou Central Hospital, Xuzhou, China
| | - Yue-Yue Liu
- Department of Radiology, Xuzhou Central Hospital, Xuzhou, China
| | - Tao Wang
- Department of Radiology, Xuzhou Central Hospital, Xuzhou, China
| | - Ya-Yong Huang
- Department of Radiology, Xuzhou Central Hospital, Xuzhou, China
| | - Ping Xia
- Department of Radiology, Xuzhou Central Hospital, Xuzhou, China.
| |
Collapse
|
7
|
Duignan JA, Ryan DT, O'Riordan B, O'Brien A, Healy GM, O'Brien C, Butler M, Keane MP, McCarthy C, Murphy DJ, Dodd JD. Combined autologous blood patch-immediate patient rollover does not reduce the pneumothorax or chest drain rate following CT-guided lung biopsy compared to immediate patient rollover alone. Eur J Radiol 2023; 160:110691. [PMID: 36640713 DOI: 10.1016/j.ejrad.2023.110691] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/14/2022] [Revised: 12/27/2022] [Accepted: 01/07/2023] [Indexed: 01/11/2023]
Abstract
PUPROSE The purpose of this study was to evaluate a combined autologous blood-patch (ABP)-immediate patient rollover (IPR) technique compared with the IPR technique alone on the incidence of pneumothorax and chest drainage following CT-guided lung biopsy. METHODS In this interventional cohort study of both prospectively and retrospectively acquired data, 652 patients underwent CT-guided lung biopsy. Patient demographics, lesion characteristics and technical biopsy variables including the combined ABP-IPR versus IPR alone were evaluated as predictors of pneumothorax and chest drain rates using regression analysis. RESULTS The combined ABP-IPR technique was performed in 259 (39.7 %) patients whilst 393 (60.3 %) underwent IPR alone. There was no significant difference in pneumothorax rate or chest drains required between the combined ABP-IPR vs IPR groups (p =.08, p =.60 respectively). Predictors of pneumothorax adjusted for the combined ABP-IPR and IPR alone groups included age (p =.02), lesion size (p =.01), location (p =.005), patient position (p =.008), emphysema along the needle track (p =.005) and lesion distance from the pleura (p =.02). Adjusted predictors of chest drain insertion included lesion location (p =.09), patient position (p =.002), bullae crossed (p =.02) and lesion distance from the pleura (p =.02). CONCLUSION The combined ABP-IPR technique does not reduce the pneumothorax or chest drain rate compared to the IPR technique alone. Utilising IPR without an ABP following CT-guided lung biopsy results in similar pneumothorax and chest drain rates while minimising the potential risk of systemic air embolism.
Collapse
Affiliation(s)
- John A Duignan
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland.
| | - David T Ryan
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland
| | - Brian O'Riordan
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland; School of Medicine, University College Dublin, Ireland
| | - Amy O'Brien
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland
| | - Gerard M Healy
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland
| | - Cormac O'Brien
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland
| | - Marcus Butler
- Department of Respiratory Medicine, St. Vincent's University Hospital, Dublin, Ireland; School of Medicine, University College Dublin, Ireland
| | - Michael P Keane
- Department of Respiratory Medicine, St. Vincent's University Hospital, Dublin, Ireland; School of Medicine, University College Dublin, Ireland
| | - Cormac McCarthy
- Department of Respiratory Medicine, St. Vincent's University Hospital, Dublin, Ireland; School of Medicine, University College Dublin, Ireland
| | - David J Murphy
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland; School of Medicine, University College Dublin, Ireland
| | - Jonathan D Dodd
- Department of Radiology, St. Vincent's University Hospital, Dublin, Ireland; School of Medicine, University College Dublin, Ireland.
| |
Collapse
|
8
|
Clinical value of CT-guided biopsy of small (≤1.5 cm) suspicious lung nodules: Diagnostic accuracy, molecular characterization and long-term clinical outcomes. Cancer Treat Res Commun 2022; 33:100626. [PMID: 36041372 DOI: 10.1016/j.ctarc.2022.100626] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/24/2022] [Revised: 08/10/2022] [Accepted: 08/20/2022] [Indexed: 12/14/2022]
Abstract
Small pulmonary nodules (≤1.5 cm) are frequently detected on routine chest imaging and lung cancer screening studies. Our goal was to determine the clinical value of CT-guided core needle biopsy (CNB) in the evaluation of such nodules. In this single-center study, we retrospectively analyzed patient data (n = 44) for CNBs on lung nodules (≤1.5 cm) performed at our biopsy center between May 2017 and March 2020. We analyzed for the rate of pathology diagnosis, molecular/biomarker analysis, complications, and change in clinical management and outcome over a period ranging up to 60 months after biopsy. A pathology diagnosis of malignancy or benign lesion was obtained in 97.9% of biopsies in this cohort. The rate of complications was low with only 6.8% of patients requiring the insertion of a temporary small profile interventional radiology (IR) pigtail chest tube for pneumothorax. Out of the subset of biopsy specimens that were sent for tissue molecular analysis, 90% had enough tissue preserved after initial pathological analysis to obtain at least one molecular marker. Our data show that CT-guided CNB is safe and reliable, and should be considered for the evaluation of small, suspicious lung nodules found on routine screenings for the early detection and evaluation of malignant lesions.
Collapse
|
9
|
Schiebler ML, Hartung M. Postprocedural Pneumothorax Detection by Deep Learning on Chest Radiographs. Radiology 2022; 303:442-443. [PMID: 35076307 DOI: 10.1148/radiol.212973] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/11/2022]
Affiliation(s)
- Mark L Schiebler
- From the Department of Radiology, UW-Madison School of Medicine and Public Health, 600 Highland Ave, E3/378 Clinical Science Center, Madison, WI 53794
| | - Michael Hartung
- From the Department of Radiology, UW-Madison School of Medicine and Public Health, 600 Highland Ave, E3/378 Clinical Science Center, Madison, WI 53794
| |
Collapse
|