Oliver Pech
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Oliver Pech has a total of 231 co-authors. Here are the top co-authors by number of shared publications:
- Christian Ell
Departments of Internal Medicine II, Sana Hospital, Offenbach, Germany
16 shared publications - Andrea May
Klinikum Offenbach, Offenbach, Germany
13 shared publications - Hendrik Manner
Department of Internal Medicine II, HSK Hospital, Teaching Hospital of the University Medicine of Mainz, Ludwig-Erhard-Strasse 100, 65199, Wiesbaden, Germany.
12 shared publications - Raf Bisschops
University Hospitals Leuven, KU Leuven, Belgium
7 shared publications C. Ell
6 shared publicationsChristian Ell
HSK Wiesbaden (Teaching Hospital of the University of Mainz), Wiesbaden, Germany
6 shared publications
Latest Publications
- Open Access
- 26 November 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure from eQTL Set–Based Association Analysis Identifies Novel Susceptibility Loci for Barrett Esophagus and Esophageal Adenocarcinoma
Supplementary Figure from eQTL Set–Based Association Analysis Identifies Novel Susceptibility Loci for Barrett Esophagus and Esophageal Adenocarcinoma
- 26 November 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure from eQTL Set–Based Association Analysis Identifies Novel Susceptibility Loci for Barrett Esophagus and Esophageal Adenocarcinoma
Supplementary Figure from eQTL Set–Based Association Analysis Identifies Novel Susceptibility Loci for Barrett Esophagus and Esophageal Adenocarcinoma
- Open Access
Clinical consequences of computer-aided colorectal polyp detection
Background and aim Randomised trials show improved polyp detection with computer-aided detection (CADe), mostly of small lesions. However, operator and selection bias may affect CADe’s true benefit. Clinical outcomes of increased detection have not yet been fully elucidated. Methods In this multicentre trial, CADe combining convolutional and recurrent neural networks was used for polyp detection. Blinded endoscopists were monitored in real time by a second observer with CADe access. CADe detections prompted reinspection. Adenoma detection rates (ADR) and polyp detection rates were measured prestudy and poststudy. Histological assessments were done by independent histopathologists. The primary outcome compared polyp detection between endoscopists and CADe. Results In 946 patients (51.9% male, mean age 64), a total of 2141 polyps were identified, including 989 adenomas. CADe was not superior to human polyp detection (sensitivity 94.6% vs 96.0%) but outperformed them when restricted to adenomas. Unblinding led to an additional yield of 86 true positive polyp detections (1.1% ADR increase per patient; 73.8% were Conclusion Even if CADe appears to be slightly more sensitive than human endoscopists, the additional gain in ADR was minimal and follow-up intervals rarely changed. Additional inspection of non-neoplastic lesions was increased, adding to the inspection and/or polypectomy workload.
Open AccessClinical consequences of computer-aided colorectal polyp detection
Background and aim Randomised trials show improved polyp detection with computer-aided detection (CADe), mostly of small lesions. However, operator and selection bias may affect CADe’s true benefit. Clinical outcomes of increased detection have not yet been fully elucidated. Methods In this multicentre trial, CADe combining convolutional and recurrent neural networks was used for polyp detection. Blinded endoscopists were monitored in real time by a second observer with CADe access. CADe detections prompted reinspection. Adenoma detection rates (ADR) and polyp detection rates were measured prestudy and poststudy. Histological assessments were done by independent histopathologists. The primary outcome compared polyp detection between endoscopists and CADe. Results In 946 patients (51.9% male, mean age 64), a total of 2141 polyps were identified, including 989 adenomas. CADe was not superior to human polyp detection (sensitivity 94.6% vs 96.0%) but outperformed them when restricted to adenomas. Unblinding led to an additional yield of 86 true positive polyp detections (1.1% ADR increase per patient; 73.8% were Conclusion Even if CADe appears to be slightly more sensitive than human endoscopists, the additional gain in ADR was minimal and follow-up intervals rarely changed. Additional inspection of non-neoplastic lesions was increased, adding to the inspection and/or polypectomy workload.
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S246
Analysis of deeper sections from initial false-negative diagnosis after cold snare polypectomy of small adenomas <10mm
Aims While performing a realtime CADx study a negative histopathological diagnosis was obtained in a portion of specimens from lesions detected as adenoma by CADx and/or endoscopists. In the present study, we asked, if small adenomatous lesion can be found in the deeper section within the remaining embedded tissue. Methods The polyps (cold snare polypectomy) in our CADx study were sent to a local pathology institute. Two experienced pathologists reviewed the histopathological slides independently. Standard work up is cutting 6-8 sections of the paraffin block per polyp. In our prospective CADx study cohort of 262 polyps, 34 lesions were diagnosed as non-adenomas on routine histology while AI and/or endoscopists diagnosed “adenoma”. In a second step, the paraffin blocks of these polyps were completely cut in multiple serial sections. The section with most adenomatous tissue was described with number and depth of positive crypts. Results 34 polyps (27 patients) were re-evaluated as described. In 11 (32.4%) of 34 polyps the final diagnosis changed from non-adenomatous to tubular adenoma (LGIN) and one lesion (2.9%) to sessile serrated polyp. The number of positive crypts varied from 5-70 (mean 21.42 SD 18,63), the depth of positive crypts from 20-100% (mean 65, SD 29.69). Remaining non-adenomas were 18 (52.9%) hyperplastic polyps, 2 (5.9%) lymphoid aggregates, 1 (2.9%) non-adenomatous colon mucosa, and 1 (2.9%) colonic ganglioneuroma. Conclusions A relevant portion of adenomas was missed on standard histopathological work up. Cutting deeper sections in cases with disconcordant diagnosis is inevitable to correct final diagnosis. This might have an impact on the recommended surveillance intervals after polypectomy. Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S246
Analysis of deeper sections from initial false-negative diagnosis after cold snare polypectomy of small adenomas <10mm
Aims While performing a realtime CADx study a negative histopathological diagnosis was obtained in a portion of specimens from lesions detected as adenoma by CADx and/or endoscopists. In the present study, we asked, if small adenomatous lesion can be found in the deeper section within the remaining embedded tissue. Methods The polyps (cold snare polypectomy) in our CADx study were sent to a local pathology institute. Two experienced pathologists reviewed the histopathological slides independently. Standard work up is cutting 6-8 sections of the paraffin block per polyp. In our prospective CADx study cohort of 262 polyps, 34 lesions were diagnosed as non-adenomas on routine histology while AI and/or endoscopists diagnosed “adenoma”. In a second step, the paraffin blocks of these polyps were completely cut in multiple serial sections. The section with most adenomatous tissue was described with number and depth of positive crypts. Results 34 polyps (27 patients) were re-evaluated as described. In 11 (32.4%) of 34 polyps the final diagnosis changed from non-adenomatous to tubular adenoma (LGIN) and one lesion (2.9%) to sessile serrated polyp. The number of positive crypts varied from 5-70 (mean 21.42 SD 18,63), the depth of positive crypts from 20-100% (mean 65, SD 29.69). Remaining non-adenomas were 18 (52.9%) hyperplastic polyps, 2 (5.9%) lymphoid aggregates, 1 (2.9%) non-adenomatous colon mucosa, and 1 (2.9%) colonic ganglioneuroma. Conclusions A relevant portion of adenomas was missed on standard histopathological work up. Cutting deeper sections in cases with disconcordant diagnosis is inevitable to correct final diagnosis. This might have an impact on the recommended surveillance intervals after polypectomy. Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S301
Endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) of extraintestinal solid lesions: diagnostic value and safety in one single german center
Aims Endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) of solid extraintestinal lesions is an established standard procedure in daily routine. The goal of this study was to investigate how representative the obtained material is and how often complications occur. Methods Over the course of 36 months all patients with EUS-FNB of solid extraintestinal lesions in our teaching hospital were included.In addition to histologic and cytologic results,complications and needle size were assessed. Results A total of 118 FNBs were performed in the 116 patients included (mean age 66 years, 47% women). Of these, 79 (67%) were pancreatic lesions, 22 (19%) lymph nodes and 17 (14%) other solid extraintestinal lesions from the upper/lower gastrointestinal tract.FNB revealed representative material in 82% of pancreatic lesions, 91% of lymph nodes, and in 82% in other lesions. Final diagnosis was pancreatic carcinoma in 28% (n=33), NET in 6% (n=7), metastases in 6% (n=7), other carcinomas in 13% (n=15) and cystadenomas in 10% (n=12).Simultaneous histologic and cytologic evaluation was performed in 80 FNBs (68%). Of these, the diagnosis was made in both workups in 54% (n=43), in histology alone in 21% (n=17), and in no case in cytology alone.Complications occurred in 2 cases (one non-Hb relevant hemorrhage, one pancreatitis). On average, 1.6 FNBs were performed. Mainly a 22G needle (53%) followed by a 19G needle (21%) was used. Conclusions EUS-FNB is a good and safe procedure in the diagnosis of solid extraintestinal lesions. In our setting, cytologic workup seems dispensable if histologic examination has been performed. Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S301
Endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) of extraintestinal solid lesions: diagnostic value and safety in one single german center
Aims Endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) of solid extraintestinal lesions is an established standard procedure in daily routine. The goal of this study was to investigate how representative the obtained material is and how often complications occur. Methods Over the course of 36 months all patients with EUS-FNB of solid extraintestinal lesions in our teaching hospital were included.In addition to histologic and cytologic results,complications and needle size were assessed. Results A total of 118 FNBs were performed in the 116 patients included (mean age 66 years, 47% women). Of these, 79 (67%) were pancreatic lesions, 22 (19%) lymph nodes and 17 (14%) other solid extraintestinal lesions from the upper/lower gastrointestinal tract.FNB revealed representative material in 82% of pancreatic lesions, 91% of lymph nodes, and in 82% in other lesions. Final diagnosis was pancreatic carcinoma in 28% (n=33), NET in 6% (n=7), metastases in 6% (n=7), other carcinomas in 13% (n=15) and cystadenomas in 10% (n=12).Simultaneous histologic and cytologic evaluation was performed in 80 FNBs (68%). Of these, the diagnosis was made in both workups in 54% (n=43), in histology alone in 21% (n=17), and in no case in cytology alone.Complications occurred in 2 cases (one non-Hb relevant hemorrhage, one pancreatitis). On average, 1.6 FNBs were performed. Mainly a 22G needle (53%) followed by a 19G needle (21%) was used. Conclusions EUS-FNB is a good and safe procedure in the diagnosis of solid extraintestinal lesions. In our setting, cytologic workup seems dispensable if histologic examination has been performed. Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S10
Clinical validation of a computer-aided detection model for colorectal polyp detection (CAD-ARTIPOD) trial using a second observer and real-time unblinding
Aims This trial aimed to validate a CADe system for colorectal polyp detection versus trained endoscopists in an endoscopy blinded set-up with real-time unblinding. Methods The CAD-ARTIPOD trial is a prospective multicenter trial conducted in 9 European centers. CADe specifics were published earlier [1]. Sample size analysis for superiority required 2048 polyps assuming a difference in sensitivity of 2% with endoscopic diagnosis as gold standard. All endoscopists (ADR 20-50%) were blinded for the CADe output. A second observer classified in real time the CADe output in (1) true positive (detected/not detected by the endoscopist) (2) false negative or (3) clinically relevant or irrelevant false positive detection. If a polyp was only detected by CADe, the endoscopist was unblinded for confirmation of the finding. Histological assessment of endoscopically confirmed polyps was performed by two independent histopathologists. Results A number of 2080 polyps were collected, 1549 were histologically confirmed polyps. For the primary endpoint superiority was not reached between CADe and endoscopists (sensitivity (Sn) of 0.94 vs 0.96) (non-inferiority margin of 5%, p = 0.0216). False positive rate was 1.8/min. With histology as standard, CADe outperformed the endoscopists (Sn 0.96 versus 0.95 (p=0.03)) and showed an extra detection rate of 4.8% (Fig. [1]). Conclusions This manufacturer independent CADe system used in an innovative and competitive study design demonstrated non-inferiority in comparison to endoscopists’ polyp detection. However, with histology as gold standard CADe outperformed endoscopists. Our study also shows that “watching over a shoulder” in itself leads to higher PDR and ADR Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S10
Clinical validation of a computer-aided detection model for colorectal polyp detection (CAD-ARTIPOD) trial using a second observer and real-time unblinding
Aims This trial aimed to validate a CADe system for colorectal polyp detection versus trained endoscopists in an endoscopy blinded set-up with real-time unblinding. Methods The CAD-ARTIPOD trial is a prospective multicenter trial conducted in 9 European centers. CADe specifics were published earlier [1]. Sample size analysis for superiority required 2048 polyps assuming a difference in sensitivity of 2% with endoscopic diagnosis as gold standard. All endoscopists (ADR 20-50%) were blinded for the CADe output. A second observer classified in real time the CADe output in (1) true positive (detected/not detected by the endoscopist) (2) false negative or (3) clinically relevant or irrelevant false positive detection. If a polyp was only detected by CADe, the endoscopist was unblinded for confirmation of the finding. Histological assessment of endoscopically confirmed polyps was performed by two independent histopathologists. Results A number of 2080 polyps were collected, 1549 were histologically confirmed polyps. For the primary endpoint superiority was not reached between CADe and endoscopists (sensitivity (Sn) of 0.94 vs 0.96) (non-inferiority margin of 5%, p = 0.0216). False positive rate was 1.8/min. With histology as standard, CADe outperformed the endoscopists (Sn 0.96 versus 0.95 (p=0.03)) and showed an extra detection rate of 4.8% (Fig. [1]). Conclusions This manufacturer independent CADe system used in an innovative and competitive study design demonstrated non-inferiority in comparison to endoscopists’ polyp detection. However, with histology as gold standard CADe outperformed endoscopists. Our study also shows that “watching over a shoulder” in itself leads to higher PDR and ADR Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S308-S309
ERCP in infants, children and adolescents – Experiences of a pediatric centre
Aims ERCP in children older than 3 years and adolescents is a common procedure for detecting and treating congenital or acquired biliopancreatic disorders. For infants younger than 3 months ERCP represents a complex and challenging technique to diagnose or treat rare, mostly congenital diseases. Still there is poor data regarding diagnostic yield, therapeutic outcome and above all the savety of ERCP in infants younger than 3 months. Methods To investigate this, we conducted a retrospective analysis of 90 ERCPs in 63 children between Jan. 2019 and Dec. 2021. Aim was to assess the indication for ERCPs, the diagnostic and therapeutic outcome as well as the rate of adverse events during and after the procedures. Results Over a course of two years a total of 63 consecutive children (mean age 2.88 years, median 0.17 years, 45% girls) underwent 90 ERCPs. 37 children (58%) were younger than 3 months, 28 children (29%) were between 3 months and 6 years and 8 children (13%) were older than 6 years. The main reasons for ERCPs in infants were congenital disorders, such as bile duct astresia or hypoplasia. In children and adolescents ERCP was indicated because of congenital and acquired disorders (gallstones, hereditary pancreatitis). The mean duration of ERCPs overall was 40.7 minutes (diagnostic 30.3 minutes, therapeutic 51.1 minutes). Complications occured in 7 cases (7.7%): 1 post ERCP pancreatitis, 4 mild cholangitis, 1 endotracheal re-intubation due to respiratory insufficiency after procedure [1] [2] [3] [4] [5] [6]. Conclusions In this large retrospective cohort study we were able to show that pediatric ERCP is a feasible, effective and save procedure, even in infants younger than 3 months. Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 14 April 2023
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 55 (S 02) , S308-S309
ERCP in infants, children and adolescents – Experiences of a pediatric centre
Aims ERCP in children older than 3 years and adolescents is a common procedure for detecting and treating congenital or acquired biliopancreatic disorders. For infants younger than 3 months ERCP represents a complex and challenging technique to diagnose or treat rare, mostly congenital diseases. Still there is poor data regarding diagnostic yield, therapeutic outcome and above all the savety of ERCP in infants younger than 3 months. Methods To investigate this, we conducted a retrospective analysis of 90 ERCPs in 63 children between Jan. 2019 and Dec. 2021. Aim was to assess the indication for ERCPs, the diagnostic and therapeutic outcome as well as the rate of adverse events during and after the procedures. Results Over a course of two years a total of 63 consecutive children (mean age 2.88 years, median 0.17 years, 45% girls) underwent 90 ERCPs. 37 children (58%) were younger than 3 months, 28 children (29%) were between 3 months and 6 years and 8 children (13%) were older than 6 years. The main reasons for ERCPs in infants were congenital disorders, such as bile duct astresia or hypoplasia. In children and adolescents ERCP was indicated because of congenital and acquired disorders (gallstones, hereditary pancreatitis). The mean duration of ERCPs overall was 40.7 minutes (diagnostic 30.3 minutes, therapeutic 51.1 minutes). Complications occured in 7 cases (7.7%): 1 post ERCP pancreatitis, 4 mild cholangitis, 1 endotracheal re-intubation due to respiratory insufficiency after procedure [1] [2] [3] [4] [5] [6]. Conclusions In this large retrospective cohort study we were able to show that pediatric ERCP is a feasible, effective and save procedure, even in infants younger than 3 months. Article published online: 14 April 2023 © 2023. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 16 June 2022
- Published by American Association for Cancer Research (AACR) in Cancer Epidemiology, Biomarkers & Prevention
- Vol. 31 (9) , 1735-1745
eQTL Set–Based Association Analysis Identifies Novel Susceptibility Loci for Barrett Esophagus and Esophageal Adenocarcinoma
Background: Over 20 susceptibility single-nucleotide polymorphisms (SNP) have been identified for esophageal adenocarcinoma (EAC) and its precursor, Barrett esophagus (BE), explaining a small portion of heritability. Methods: Using genetic data from 4,323 BE and 4,116 EAC patients aggregated by international consortia including the Barrett's and Esophageal Adenocarcinoma Consortium (BEACON), we conducted a comprehensive transcriptome-wide association study (TWAS) for BE/EAC, leveraging Genotype Tissue Expression (GTEx) gene-expression data from six tissue types of plausible relevance to EAC etiology: mucosa and muscularis from the esophagus, gastroesophageal (GE) junction, stomach, whole blood, and visceral adipose. Two analytical approaches were taken: standard TWAS using the predicted gene expression from local expression quantitative trait loci (eQTL), and set-based SKAT association using selected eQTLs that predict the gene expression. Results: Although the standard approach did not identify significant signals, the eQTL set–based approach identified eight novel associations, three of which were validated in independent external data (eQTL SNP sets for EXOC3, ZNF641, and HSP90AA1). Conclusions: This study identified novel genetic susceptibility loci for EAC and BE using an eQTL set–based genetic association approach. Impact: This study expanded the pool of genetic susceptibility loci for EAC and BE, suggesting the potential of the eQTL set–based genetic association approach as an alternative method for TWAS analysis.
- 16 June 2022
- Published by American Association for Cancer Research (AACR) in Cancer Epidemiology, Biomarkers & Prevention
- Vol. 31 (9) , 1735-1745
eQTL Set–Based Association Analysis Identifies Novel Susceptibility Loci for Barrett Esophagus and Esophageal Adenocarcinoma
Background: Over 20 susceptibility single-nucleotide polymorphisms (SNP) have been identified for esophageal adenocarcinoma (EAC) and its precursor, Barrett esophagus (BE), explaining a small portion of heritability. Methods: Using genetic data from 4,323 BE and 4,116 EAC patients aggregated by international consortia including the Barrett's and Esophageal Adenocarcinoma Consortium (BEACON), we conducted a comprehensive transcriptome-wide association study (TWAS) for BE/EAC, leveraging Genotype Tissue Expression (GTEx) gene-expression data from six tissue types of plausible relevance to EAC etiology: mucosa and muscularis from the esophagus, gastroesophageal (GE) junction, stomach, whole blood, and visceral adipose. Two analytical approaches were taken: standard TWAS using the predicted gene expression from local expression quantitative trait loci (eQTL), and set-based SKAT association using selected eQTLs that predict the gene expression. Results: Although the standard approach did not identify significant signals, the eQTL set–based approach identified eight novel associations, three of which were validated in independent external data (eQTL SNP sets for EXOC3, ZNF641, and HSP90AA1). Conclusions: This study identified novel genetic susceptibility loci for EAC and BE using an eQTL set–based genetic association approach. Impact: This study expanded the pool of genetic susceptibility loci for EAC and BE, suggesting the potential of the eQTL set–based genetic association approach as an alternative method for TWAS analysis.
- Open Access
- 12 May 2022
- Published by S. Karger AG in Visceral Medicine
- Vol. 38 (3) , 159-160
Barrett’s Esophagus: Today’s Mistakes and Tomorrow’s Wisdom
- 12 May 2022
- Published by S. Karger AG in Visceral Medicine
- Vol. 38 (3) , 159-160
Open AccessBarrett’s Esophagus: Today’s Mistakes and Tomorrow’s Wisdom
- 27 January 2022
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 54 (02) , 223
Reply to Chen et al.
We would like to thank Chen et al. for their comments on our study [1]. The artificial intelligence (AI) system performed comparably to experienced endoscopists. Compared with larger trials that evaluated the same AI system (GI Genius; Medtronic, Minneapolis, Minnesota, USA), the adenoma detection rates (ADRs) of the computer-aided detection groups [2] [3] were similar to the ADR of the AI system in our study. This study was conducted to compare the AI with experienced endoscopists, who served as the control group. It was not designed to evaluate the impact of the AI on the performance of individual endoscopists. Therefore, endoscopists were not divided into subgroups. We agree that for further evaluation, endoscopists should be divided into subgroups with larger sample sizes. As the endoscopy image was simultaneously displayed on two opposing monitors, endoscopists and AI had equal chances to detect a polyp on their monitor, regardless of the time operators needed for observation and manipulation. Transient detections of the AI were re-evaluated by endoscopists. Observation time of endoscopists was not decreased due to the study design. As the monitors of the AI and the endoscopists simultaneously displayed the same endoscopy image, and as withdrawal was paused for resection, no advantages were given to the AI system. AI was not switched off for resection as this would have elongated the colonoscopy procedure without benefits. In addition, we did not take biopsies of all detections of the AI as this would have been of no benefit to the patient. Every detection of the AI was carefully re-evaluated by the endoscopists, whose decision was used as the criterion standard. Publication Date: 27 January 2022 (online) Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
- 27 January 2022
- Published by Georg Thieme Verlag KG in Endoscopy
- Vol. 54 (02) , 223
Reply to Chen et al.
We would like to thank Chen et al. for their comments on our study [1]. The artificial intelligence (AI) system performed comparably to experienced endoscopists. Compared with larger trials that evaluated the same AI system (GI Genius; Medtronic, Minneapolis, Minnesota, USA), the adenoma detection rates (ADRs) of the computer-aided detection groups [2] [3] were similar to the ADR of the AI system in our study. This study was conducted to compare the AI with experienced endoscopists, who served as the control group. It was not designed to evaluate the impact of the AI on the performance of individual endoscopists. Therefore, endoscopists were not divided into subgroups. We agree that for further evaluation, endoscopists should be divided into subgroups with larger sample sizes. As the endoscopy image was simultaneously displayed on two opposing monitors, endoscopists and AI had equal chances to detect a polyp on their monitor, regardless of the time operators needed for observation and manipulation. Transient detections of the AI were re-evaluated by endoscopists. Observation time of endoscopists was not decreased due to the study design. As the monitors of the AI and the endoscopists simultaneously displayed the same endoscopy image, and as withdrawal was paused for resection, no advantages were given to the AI system. AI was not switched off for resection as this would have elongated the colonoscopy procedure without benefits. In addition, we did not take biopsies of all detections of the AI as this would have been of no benefit to the patient. Every detection of the AI was carefully re-evaluated by the endoscopists, whose decision was used as the criterion standard. Publication Date: 27 January 2022 (online) Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany