1
|
Georgiou MF, Sfakianaki E, Diaz-Kanelidis MN, Moshiree B. Gastric Emptying Scintigraphy Protocol Optimization Using Machine Learning for the Detection of Delayed Gastric Emptying. Diagnostics (Basel) 2024; 14:1240. [PMID: 38928655 PMCID: PMC11202747 DOI: 10.3390/diagnostics14121240] [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: 02/22/2024] [Revised: 06/01/2024] [Accepted: 06/10/2024] [Indexed: 06/28/2024] Open
Abstract
PURPOSE The purpose of this study is to examine the feasibility of a machine learning (ML) system for optimizing a gastric emptying scintigraphy (GES) protocol for the detection of delayed gastric emptying (GE), which is considered a primary indication for the diagnosis of gastroparesis. METHODS An ML model was developed using the JADBio AutoML artificial intelligence (AI) platform. This model employs the percent GE at various imaging time points following the ingestion of a standardized radiolabeled meal to predict normal versus delayed GE at the conclusion of the 4 h GES study. The model was trained and tested on a cohort of 1002 patients who underwent GES using a 70/30 stratified split ratio for training vs. testing. The ML software automated the generation of optimal predictive models by employing a combination of data preprocessing, appropriate feature selection, and predictive modeling analysis algorithms. RESULTS The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was employed to evaluate the predictive modeling performance. Several models were developed using different combinations of imaging time points as input features and methodologies to achieve optimal output. By using GE values at time points 0.5 h, 1 h, 1.5 h, 2 h, and 2.5 h as input predictors of the 4 h outcome, the analysis produced an AUC of 90.7% and a balanced accuracy (BA) of 80.0% on the test set. This performance was comparable to the training set results (AUC = 91.5%, BA = 84.7%) within the 95% confidence interval (CI), demonstrating a robust predictive capability. Through feature selection, it was discovered that the 2.5 h GE value alone was statistically significant enough to predict the 4 h outcome independently, with a slightly increased test set performance (AUC = 92.4%, BA = 83.3%), thus emphasizing its dominance as the primary predictor for delayed GE. ROC analysis was also performed for single time imaging points at 1 h and 2 h to assess their independent predictiveness of the 4 h outcome. Furthermore, the ML model was tested for its ability to predict "flipping" cases with normal GE at 1 h and 2 h that became abnormal with delayed GE at 4 h. CONCLUSIONS An AI/ML model was designed and trained for predicting delayed GE using a limited number of imaging time points in a 4 h GES clinical protocol. This study demonstrates the feasibility of employing ML for GES optimization in the detection of delayed GE and potentially shortening the protocol's time length without compromising diagnostic power.
Collapse
Affiliation(s)
- Michalis F. Georgiou
- Department of Radiology, University of Miami Miller School of Medicine, Miami, FL 33136, USA;
| | - Efrosyni Sfakianaki
- Department of Radiology, University of Miami Miller School of Medicine, Miami, FL 33136, USA;
| | | | - Baha Moshiree
- Atrium Health, Wake Forest University, Charlotte, NC 28204, USA
| |
Collapse
|
2
|
Early steps of an alternative test meal for gastric emptying scintigraphy. VOJNOSANIT PREGL 2022. [DOI: 10.2298/vsp211121052s] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/27/2022] Open
Abstract
Background/Aim. Gastric emptying (GE) scintigraphy provides a physiologic and noninvasive measurement of GE. Although GE scintigraphy has been standardized, preparing a meal is still complex and not practical in daily routine. The aim of the study was to prepare a simple, practical, and easily standardizable semisolid meal and investigate its role in estimating the GE function in of rabbits. Methods. In the first part of the study (basal condition), the mixture of the macroaggregated albumin (MAA) labeled with 37 MBq (1 mCi) of technetium-99m (99mTc) and 40 g of barium sulfate (1g/mL) was applied to animals via a nasogastric catheter. A series of images (frame/min, 60 min) in the anterior and posterior projections were dynamically acquired, and the motion was corrected after the radiopharmaceutical application. A few days later, the same rabbits were scanned under the same protocol after a 1 mg atropine injection to simulate gastroparesis condition. Eleven rabbits were included according to inclusion and exclusion criteria, and a total of twenty-two imaging data sets were analyzed for quantification. Results. In the basal study, total counts of the mixture decreased from 87,800.83 ? 12,622.76 to 42,733.14 ? 6,591.53 at 30 min and to 13,684.19 ? 1,774.90 at 60 min, and these decreases were statistically significant (p = 0.003). Emptying percentages were 51.39 ? 0.78% at 30 min and 84.32 ? 1.56 at 60 min and were statistically significant (p = 0.003). After intravascular atropine sulfate injection, total counts of the mixture decreased from 84,508.78 ? 11,871.48 to 64,995.18 ? 9,298 at 30 min and to 53,507.17 ? 7,258.98 at 60 min, and these decreases were statistically significant (p = 0.003). Emptying percentages were 23.10 ? 1.11% at 30 min and 36.63 ? 1.42 at 60 min and were statistically significant (p = 0.003). The difference between basal and post-atropine sulfate gastric emptying percentage at 30th (p = 0.003) and 60th (p = 0.003) min was statistically significant. Conclusion. The meal, used in this study, is non-nutrient, fatty-free, and semisolid and is easy to prepare and administer. Due to its semisolid nature, it offers a chance to evaluate the quantification of regional and total GE as well as the separate roles of the fundus and antrum.
Collapse
|
3
|
Singh S, McCrary J, Kedar A, Weeks S, Beauerle B, Weeks A, Endashaw O, Lahr C, Starkebaum W, Abell T. Temporary Endoscopic Stimulation in Gastroparesis-like Syndrome. J Neurogastroenterol Motil 2015; 21:520-7. [PMID: 26351250 PMCID: PMC4622134 DOI: 10.5056/jnm15046] [Citation(s) in RCA: 19] [Impact Index Per Article: 2.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/15/2015] [Revised: 05/27/2015] [Accepted: 06/23/2015] [Indexed: 01/16/2023] Open
Abstract
Background/Aims Gastroparesis-like syndrome (GLS) is defined as gastroparesis-like symptoms with normal gastric scintigraphy. While the efficacy of gastric electrical stimulation (GES) in gastroparesis is well known, the utility of GES in GLS is largely unknown. Our aim was to clarify the role of GES in GLS. We implanted consecutive patients with symptoms of gastroparesis with temporary gastric electrical stimulation and observed changes in gastric scintigraphy and total symptom score. Methods Five hundred and fifty-one patients suffering from symptoms of gastroparesis (nausea, vomiting, bloating/distension, anorexia/early satiety, and abdominal pain) with negative endoscopy underwent gastric scintigraphy with analysis of 1) solid radio-nuclide gastric emptying at 1, 2, and 4 hours (% remaining); 2) area under the gastric emptying curve (AUC) at 1, 2, and 4 hours; and 3) total gastric emptying test (GET) (the sum of 1, 2, and 4 hour values). Patients were stratified into: delayed gastric emptying, normal gastric emptying, and rapid gastric emptying (Appendix). Of the 551 patients in the larger cohort, 379 had implantation of temporary gastric electrical stimulation (tGES). Gastrointestinal symptoms and gastric emptying were com -pared pre and post tGES implantation. Results After tGES, 2 hour gastric retention decreased (P < 0.01) for the delayed patients, and increased (P < 0.001) for normal and rapid patients. These changes were accompanied by improvements (P < 0.001) in vomiting, nausea, and total symptom scores in all 3 subgroups. Conclusions Gastric electrical stimulation may be an effective therapy for treating the symptoms of gastroparesis with normal gastric emptying. Further exploration of endoscopic electrical stimulation as a treatment for gastroparesis-like symptoms with non-delayed gastric emptying is needed.
Collapse
Affiliation(s)
- Sanjeev Singh
- Department of General Internal Medicine, Palliative Medicine and Medical Education, Department of Medicine, University of Louisville School of Medicine, Louisville, Kentucky, USA.,Division of Gastroenterology, Hepatology & Nutrition, Department of Medicine, University of Louisville School of Medicine, Louisville, Kentucky, USA
| | - Jeff McCrary
- Division of Gastroenterology, Department of Medicine, University of Mississippi Medical Center, Jackson, Mississippi, USA
| | - Archana Kedar
- Division of Gastroenterology, Hepatology & Nutrition, Department of Medicine, University of Louisville School of Medicine, Louisville, Kentucky, USA
| | - Stephen Weeks
- Division of Gastroenterology, Department of Medicine, University of Mississippi Medical Center, Jackson, Mississippi, USA
| | - Brian Beauerle
- Division of Gastroenterology, Hepatology & Nutrition, Department of Medicine, University of Louisville School of Medicine, Louisville, Kentucky, USA
| | - Andrew Weeks
- Division of Gastroenterology, Department of Medicine, University of Mississippi Medical Center, Jackson, Mississippi, USA
| | - Omer Endashaw
- Division of Gastroenterology, Hepatology & Nutrition, Department of Medicine, University of Louisville School of Medicine, Louisville, Kentucky, USA
| | - Chris Lahr
- Department of Surgery, University of Mississippi Medical Center, Jackson, Mississippi, USA
| | | | - Thomas Abell
- Department of General Internal Medicine, Palliative Medicine and Medical Education, Department of Medicine, University of Louisville School of Medicine, Louisville, Kentucky, USA.,Division of Gastroenterology, Hepatology & Nutrition, Department of Medicine, University of Louisville School of Medicine, Louisville, Kentucky, USA
| |
Collapse
|
4
|
Gastroparesis in children: the benefit of conducting 4-hour scintigraphic gastric-emptying studies. J Pediatr Gastroenterol Nutr 2013; 56:439-42. [PMID: 23111764 DOI: 10.1097/mpg.0b013e31827a789c] [Citation(s) in RCA: 19] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/10/2022]
Abstract
BACKGROUND AND AIM Scintigraphic gastric emptying study (GES) is the criterion standard for diagnosis of gastroparesis. Adult studies demonstrated that extending GES to 4 hours increases its ability to diagnose delayed gastric emptying. Most pediatric centers assess GES up to 2 hours postmeal. The aim of the present study was to assess the effect of extending GES from 2 to 4 hours in evaluation of children with suspected gastroparesis. METHODS We conducted a chart review of all children who had a 4-hour GES with standard radiolabeled solid meal in 2009-2010. Results of GES at 1, 2, and 4 hours were compared. Patients were diagnosed as having gastroparesis using adult criteria: if gastric retention of meal was >90%, 60%, and 10% at 1, 2 and 4 hours, respectively. A telephone survey assessed GES time at top 20 pediatric gastroenterology centers in the United States. Cost of evaluation of patients diagnosed as having gastroparesis was estimated. Full-time equivalents of nuclear medicine technicians and number of nuclear medicine studies done at Ann & Robert H. Lurie Children's Hospital of Chicago from 2007 to 2010 were examined. RESULTS A total of 71 patients (32 boys, average age 10.8 years) were studied. Sixty-two percent (n=44) children had abnormal GES; 23% (8/35) of them who had normal values at 2 hours had abnormal GES at 4 hours (P<0.0001). Twenty-eight percent of patients had delayed GES at 1 hour: all persisted to have abnormal GES at 2 and 4 hours. Cost of evaluation of a child for gastroparesis was $9014. Only 5 of the top 20 pediatric gastroenterology centers in the United States conducted 4-hour GES. Transitioning from 2 hours to 4 hours only required scheduling adjustments and did not result in limitation in the number of scheduled patients. CONCLUSIONS Extending GES to 4 hours results in a considerable increase in diagnosis of gastroparesis.
Collapse
|
5
|
Lin Z, Kahrilas PJ, Roman S, Boris L, Carlson D, Pandolfino JE. Refining the criterion for an abnormal Integrated Relaxation Pressure in esophageal pressure topography based on the pattern of esophageal contractility using a classification and regression tree model. Neurogastroenterol Motil 2012; 24:e356-63. [PMID: 22716041 PMCID: PMC3616504 DOI: 10.1111/j.1365-2982.2012.01952.x] [Citation(s) in RCA: 65] [Impact Index Per Article: 5.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 01/03/2023]
Abstract
BACKGROUND The Integrated Relaxation Pressure (IRP) is the esophageal pressure topography (EPT) metric used for assessing the adequacy of esophagogastric junction (EGJ) relaxation in the Chicago Classification of motility disorders. However, because the IRP value is also influenced by distal esophageal contractility, we hypothesized that its normal limits should vary with different patterns of contractility. METHODS Five hundred and twenty two selected EPT studies were used to compare the accuracy of alternative analysis paradigms to that of a motility expert (the 'gold standard'). Chicago Classification metrics were scored manually and used as inputs for MATLAB™ programs that utilized either strict algorithm-based interpretation (fixed abnormal IRP threshold of 15 mmHg) or a classification and regression tree (CART) model that selected variable IRP thresholds depending on the associated esophageal contractility. KEY RESULTS The sensitivity of the CART model for achalasia (93%) was better than that of the algorithm-based approach (85%) on account of using variable IRP thresholds that ranged from a low value of >10 mmHg to distinguish type I achalasia from absent peristalsis to a high value of >17 mmHg to distinguish type III achalasia from distal esophageal spasm. Additionally, type II achalasia was diagnosed solely by panesophageal pressurization without the IRP entering the algorithm. CONCLUSIONS & INFERENCES Automated interpretation of EPT studies more closely mimics that of a motility expert when IRP thresholds for impaired EGJ relaxation are adjusted depending on the pattern of associated esophageal contractility. The range of IRP cutoffs suggested by the CART model ranged from 10 to 17 mmHg.
Collapse
Affiliation(s)
- Zhiyue Lin
- Department of Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL, 60611 USA
| | - Peter J. Kahrilas
- Department of Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL, 60611 USA
| | - Sabine Roman
- Department of Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL, 60611 USA,Hospices Civils de Lyon and Lyon I University; Digestive Physiology, Lyon, France
| | - Lubomyr Boris
- Department of Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL, 60611 USA
| | - Dustin Carlson
- Department of Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL, 60611 USA
| | - John E. Pandolfino
- Department of Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL, 60611 USA
| |
Collapse
|
6
|
Allergic mastocytic gastroenteritis and colitis: an unexplained etiology in chronic abdominal pain and gastrointestinal dysmotility. Gastroenterol Res Pract 2012; 2012:950582. [PMID: 22577375 PMCID: PMC3346686 DOI: 10.1155/2012/950582] [Citation(s) in RCA: 23] [Impact Index Per Article: 1.9] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/23/2011] [Revised: 11/17/2011] [Accepted: 11/18/2011] [Indexed: 12/24/2022] Open
Abstract
Abdominal pain, bloating, early satiety, and changes in bowel habits are common presenting symptoms in individuals with functional GI disorders. Emerging data suggests that these symptoms may be associated with mast cell excess and/or mast cell instability in the GI tract. The aim of this retrospective study was to evaluate the contribution of mast cells to the aforementioned symptoms in individuals with a history of atopic disease. A retrospective chart review of individuals seen in a university GI practice was conducted and twenty-four subjects were identified. The majority had abdominal pain, early satiety, and nocturnal awakening. 66.7% and 37.5% had a history of environmental and/or food allergy. Solid gastric emptying was increased as were the mean number of mast cells reported on biopsies from the stomach, small bowel, and colon (>37/hpf) by CD117 staining. Mean whole blood histamine levels were uniformly elevated. This study suggests that in individuals with these characteristics, consideration should be given to staining their gastrointestinal biopsies for mast cells as this may provide them with relatively non-toxic but highly targeted treatment options. Allergic gastroenteritis and colitis may represent a third type of GI mast cell disorder along with mast cell activation syndrome and mastocytic enterocolitis.
Collapse
|