Jacques Hubert
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
Research Interests
Jacques Hubert has a total of 199 co-authors. Here are the top co-authors by number of shared publications:
- Jacques Felblinger
IADI (U1254) Université de Lorraine and Inserm Nancy France
7 shared publications Manuela Perez
IADI/Inserm U947, Lorraine University, Nancy, France; Department of Emergency and General Surgery, CHU Nancy, Nancy, France
6 shared publicationsCyril Perrenot
IADI/Inserm U947, Lorraine University, 54000, Nancy, France
5 shared publications- Benoît Feuillu
Department of Urology, University Hospital Center of Nancy‐Brabois, Rue du Morvan54500, Vandoeuvre‐les‐Nancy, France
4 shared publications - Christian Daul
CRAN (UMR 7039) Université de Lorraine CNRS Vandoeuvre-les-Nancy France
4 shared publications - Francisco Lopez-Tiro
School of Engineering and Sciences, Tecnológico de Monterrey, Monterrey, Nuevo León, México
4 shared publications
Latest Publications
- Open Access
- 1 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 53964
Corrections to “On the In Vivo Recognition of Kidney Stones Using Machine Learning”
Presents corrections to the paper, (Corrections to “On the In Vivo Recognition of Kidney Stones Using Machine Learning”).
- 1 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 53964
Open AccessCorrections to “On the In Vivo Recognition of Kidney Stones Using Machine Learning”
Presents corrections to the paper, (Corrections to “On the In Vivo Recognition of Kidney Stones Using Machine Learning”).
- 14 July 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 255
A metric learning approach for endoscopic kidney stone identification
- 14 July 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 255
A metric learning approach for endoscopic kidney stone identification
- Open Access
- 8 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 10736-10759
On the In Vivo Recognition of Kidney Stones Using Machine Learning
Determining the type of kidney stones allows urologists to prescribe a treatment to avoid the recurrence of renal lithiasis. An automated in-vivo image-based classification method would be an important step towards an immediate identification of the kidney stone type required as a first phase of the diagnosis. In the literature, it was shown on ex-vivo data (i.e., in very controlled scene and image acquisition conditions) that an automated kidney stone classification is indeed feasible. This pilot study compares the kidney stone recognition performances of six shallow machine learning methods and three deep-learning architectures which were tested with in-vivo images of the four most frequent urinary calculus types acquired with an endoscope during standard ureteroscopies. This contribution details the construction of an in-vivo dataset of endoscopic images with four of the most recurrent classes in clinical practice. It also describes the design, implementation, and results of the classifiers (shallow machine learning and deep learning-based methods) of kidney stones. Even if the best results were obtained by the InceptionV3 architecture (weighted accuracy, precision, recall and F1-Score of $97\pm 03\%, 97\pm 03\%, 98\pm 04\%$ , and $98\pm 03\%$ , respectively), it is also shown that choosing an appropriate colour space and texture features allows a shallow machine learning method to approach closely the performances of the most promising deep-learning methods (the XGBoost classifier led to weighted accuracy, precision, recall and F1-Score values of $96\pm 14\%$ ).
- 8 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 10736-10759
Open AccessOn the In Vivo Recognition of Kidney Stones Using Machine Learning
Determining the type of kidney stones allows urologists to prescribe a treatment to avoid the recurrence of renal lithiasis. An automated in-vivo image-based classification method would be an important step towards an immediate identification of the kidney stone type required as a first phase of the diagnosis. In the literature, it was shown on ex-vivo data (i.e., in very controlled scene and image acquisition conditions) that an automated kidney stone classification is indeed feasible. This pilot study compares the kidney stone recognition performances of six shallow machine learning methods and three deep-learning architectures which were tested with in-vivo images of the four most frequent urinary calculus types acquired with an endoscope during standard ureteroscopies. This contribution details the construction of an in-vivo dataset of endoscopic images with four of the most recurrent classes in clinical practice. It also describes the design, implementation, and results of the classifiers (shallow machine learning and deep learning-based methods) of kidney stones. Even if the best results were obtained by the InceptionV3 architecture (weighted accuracy, precision, recall and F1-Score of $97\pm 03\%, 97\pm 03\%, 98\pm 04\%$ , and $98\pm 03\%$ , respectively), it is also shown that choosing an appropriate colour space and texture features allows a shallow machine learning method to approach closely the performances of the most promising deep-learning methods (the XGBoost classifier led to weighted accuracy, precision, recall and F1-Score values of $96\pm 14\%$ ).
- 7 August 2023
- Published by Springer Nature in World Journal of Urology
- Vol. 41 (9) , 2481-2488
Comparison of surgical procedures for benign prostatic hyperplasia of medium-volume prostates: evaluation of the causes of rehospitalization from the French National Hospital Database (PMSI-MCO)
Purpose In France, transurethral resection of the prostate (TURP) is still the most commonly used surgical treatment for medium sized benign prostatic hyperplasia (BPH), but the Holmium Laser Enucleation of the Prostate (HoLEP) and laser vaporization procedures are becoming more common. For these three surgical procedures, we evaluate the initial complications, the short term (3 months) and the 4–12-month postoperative complications necessitating re-hospitalization. Methods From the French national hospital data base (PMSI-MCO), all hospitalizations for BPH treatment in 2018 were extracted. We document the complications during the initial hospitalization and any subsequent rehospitalizations during the one-year postoperative period. Results In 2018, 67,220 patients were treated for BPH: 46,242 TURP, 13,509 HoLEP and 7469 laser vaporization. Age and anticoagulation medications were similar for men treated by the three procedures, but TURP patients were more often hypertensive. Infections and hemorrhagic complications were the most common complications at the initial hospitalization: 17%, 10%, 13% for infections and 15%, 8.1%, 11% for hemorrhagic complications respectively, and TURP performed worse than the other two procedures at the initial hospitalization. During the first three months and then the subsequent nine months, there were fewer complications than initially, with little difference between the three procedures, all differences being less than 1%. Conclusion Laser vaporization techniques led to fewer complications. However, the PMSI-MCO only registers complications during hospitalizations. This study should be extended to non-hospitalized, more minor complications.
- 7 August 2023
- Published by Springer Nature in World Journal of Urology
- Vol. 41 (9) , 2481-2488
Comparison of surgical procedures for benign prostatic hyperplasia of medium-volume prostates: evaluation of the causes of rehospitalization from the French National Hospital Database (PMSI-MCO)
Purpose In France, transurethral resection of the prostate (TURP) is still the most commonly used surgical treatment for medium sized benign prostatic hyperplasia (BPH), but the Holmium Laser Enucleation of the Prostate (HoLEP) and laser vaporization procedures are becoming more common. For these three surgical procedures, we evaluate the initial complications, the short term (3 months) and the 4–12-month postoperative complications necessitating re-hospitalization. Methods From the French national hospital data base (PMSI-MCO), all hospitalizations for BPH treatment in 2018 were extracted. We document the complications during the initial hospitalization and any subsequent rehospitalizations during the one-year postoperative period. Results In 2018, 67,220 patients were treated for BPH: 46,242 TURP, 13,509 HoLEP and 7469 laser vaporization. Age and anticoagulation medications were similar for men treated by the three procedures, but TURP patients were more often hypertensive. Infections and hemorrhagic complications were the most common complications at the initial hospitalization: 17%, 10%, 13% for infections and 15%, 8.1%, 11% for hemorrhagic complications respectively, and TURP performed worse than the other two procedures at the initial hospitalization. During the first three months and then the subsequent nine months, there were fewer complications than initially, with little difference between the three procedures, all differences being less than 1%. Conclusion Laser vaporization techniques led to fewer complications. However, the PMSI-MCO only registers complications during hospitalizations. This study should be extended to non-hospitalized, more minor complications.
- Open Access
Boosting Kidney Stone Identification in Endoscopic Images Using Two-Step Transfer Learning
Knowing the cause of kidney stone formation is crucial to establish treatments that prevent recurrence. There are currently different approaches for determining the kidney stone type. However, the reference ex-vivo identification procedure can take up to several weeks, while an in-vivo visual recognition requires highly trained specialists. Machine learning models have been developed to provide urologists with an automated classification of kidney stones during an ureteroscopy; however, there is a general lack in terms of quality of the training data and methods. In this work, a two-step transfer learning approach is used to train the kidney stone classifier. The proposed approach transfers knowledge learned on a set of images of kidney stones acquired with a CCD camera (ex-vivo dataset) to a final model that classifies images from endoscopic images (ex-vivo dataset). The results show that learning features from different domains with similar information helps to improve the performance of a model that performs classification in real conditions (for instance, uncontrolled lighting conditions and blur). Finally, in comparison to models that are trained from scratch or by initializing ImageNet weights, the obtained results suggest that the two-step approach extracts features improving the identification of kidney stones in endoscopic images.
Open AccessBoosting Kidney Stone Identification in Endoscopic Images Using Two-Step Transfer Learning
Knowing the cause of kidney stone formation is crucial to establish treatments that prevent recurrence. There are currently different approaches for determining the kidney stone type. However, the reference ex-vivo identification procedure can take up to several weeks, while an in-vivo visual recognition requires highly trained specialists. Machine learning models have been developed to provide urologists with an automated classification of kidney stones during an ureteroscopy; however, there is a general lack in terms of quality of the training data and methods. In this work, a two-step transfer learning approach is used to train the kidney stone classifier. The proposed approach transfers knowledge learned on a set of images of kidney stones acquired with a CCD camera (ex-vivo dataset) to a final model that classifies images from endoscopic images (ex-vivo dataset). The results show that learning features from different domains with similar information helps to improve the performance of a model that performs classification in real conditions (for instance, uncontrolled lighting conditions and blur). Finally, in comparison to models that are trained from scratch or by initializing ImageNet weights, the obtained results suggest that the two-step approach extracts features improving the identification of kidney stones in endoscopic images.
- 18 July 2022
- Published by Springer Nature in Journal of Robotic Surgery
- Vol. 17 (2) , 519-526
Clinical and oncological outcomes of open partial nephrectomy versus robot assisted partial nephrectomy over 15 years
Partial nephrectomy (PN) is the gold standard surgical treatment for localized kidney cancer. The objective of our study was to compare clinical and perioperative outcomes of open partial nephrectomy (OPN) and robotic-assisted partial nephrectomy (RAPN). We retrospectively collected all patients who underwent PN for kidney cancer between 2007 and 2019 at two French academic urology departments. Clinical and perioperative outcomes and complications were compared between the OPN group and the RAPN group. Recurrence-free survival (RFS) and overall survival (OS) were compared using the log-rank test. We included 405 patients. The maximum follow-up time was 13.6 years in the OPN group and 7.1 years in the RAPN group. The OPN group was associated with more blood loss and longer hospital stay (respectively, 287 ml vs. 62.1 ml; p < 0.001 and 8.54 days vs. 4.96 days; p < 0.001). Ischemia time was shorter in the OPN group (11.4 min vs. 16.9 min; p < 0.001). The rate of complications during hospitalization and after discharge from hospital was higher in the OPN group (respectively, n = 51 vs. 30; p = 0.031 and n = 31 vs. 14; p < 0.001). RFS and OS were similar in both groups. In our study, RAPN has better perioperative outcomes with shorter hospital stay and less blood loss but also fewer early and late complications. However, we did not find any difference in terms of RFS and OS.
- 18 July 2022
- Published by Springer Nature in Journal of Robotic Surgery
- Vol. 17 (2) , 519-526
Clinical and oncological outcomes of open partial nephrectomy versus robot assisted partial nephrectomy over 15 years
Partial nephrectomy (PN) is the gold standard surgical treatment for localized kidney cancer. The objective of our study was to compare clinical and perioperative outcomes of open partial nephrectomy (OPN) and robotic-assisted partial nephrectomy (RAPN). We retrospectively collected all patients who underwent PN for kidney cancer between 2007 and 2019 at two French academic urology departments. Clinical and perioperative outcomes and complications were compared between the OPN group and the RAPN group. Recurrence-free survival (RFS) and overall survival (OS) were compared using the log-rank test. We included 405 patients. The maximum follow-up time was 13.6 years in the OPN group and 7.1 years in the RAPN group. The OPN group was associated with more blood loss and longer hospital stay (respectively, 287 ml vs. 62.1 ml; p < 0.001 and 8.54 days vs. 4.96 days; p < 0.001). Ischemia time was shorter in the OPN group (11.4 min vs. 16.9 min; p < 0.001). The rate of complications during hospitalization and after discharge from hospital was higher in the OPN group (respectively, n = 51 vs. 30; p = 0.031 and n = 31 vs. 14; p < 0.001). RFS and OS were similar in both groups. In our study, RAPN has better perioperative outcomes with shorter hospital stay and less blood loss but also fewer early and late complications. However, we did not find any difference in terms of RFS and OS.
- 17 March 2022
- Published by Elsevier in Transplantation Proceedings
- Vol. 54 (3) , 696-701
Minimally Invasive and Open Donor Nephrectomy: Lessons Learned From a French Multicenter Experience
- 17 March 2022
- Published by Elsevier in Transplantation Proceedings
- Vol. 54 (3) , 696-701
Minimally Invasive and Open Donor Nephrectomy: Lessons Learned From a French Multicenter Experience
- 2 February 2022
- Published by Wiley in The International Journal of Medical Robotics and Computer Assisted Surgery
- Vol. 18 (3) , e2381
Robot‐assisted radical prostatectomy training: Description of a canine cadaveric model
Background: Assess the relevance of a canine model in robot‐assisted radical prostatectomy (RARP) training.Methods: Step‐by‐step RARP was performed in five dog cadavers using a Da Vinci Si® Surgical Robot (Intuitive Surgical, Inc.). The steps were defined according to the RARP score, a validated training tool describing 17 key steps and four levels of difficulty; each step was scored to reflect the anatomical and technical similarities, realism of dissection, and face validity of the canine model compared to the human procedure.Results: Fourteen steps were performed during each procedure. Face validity was scored as high or very high for five of the nine steps of difficulty levels 1 and 2 as well as five of the eight steps of difficulty levels 3 and 4, especially nerve preservation, vesicourethral anastomosis and lymph node dissection.Conclusions: The cadaveric canine model seems to be a realistic and relevant training model for key steps of RARP.
- 2 February 2022
- Published by Wiley in The International Journal of Medical Robotics and Computer Assisted Surgery
- Vol. 18 (3) , e2381
Robot‐assisted radical prostatectomy training: Description of a canine cadaveric model
Background: Assess the relevance of a canine model in robot‐assisted radical prostatectomy (RARP) training.Methods: Step‐by‐step RARP was performed in five dog cadavers using a Da Vinci Si® Surgical Robot (Intuitive Surgical, Inc.). The steps were defined according to the RARP score, a validated training tool describing 17 key steps and four levels of difficulty; each step was scored to reflect the anatomical and technical similarities, realism of dissection, and face validity of the canine model compared to the human procedure.Results: Fourteen steps were performed during each procedure. Face validity was scored as high or very high for five of the nine steps of difficulty levels 1 and 2 as well as five of the eight steps of difficulty levels 3 and 4, especially nerve preservation, vesicourethral anastomosis and lymph node dissection.Conclusions: The cadaveric canine model seems to be a realistic and relevant training model for key steps of RARP.
- 1 September 2021
- Published by SAGE Publications in Journal of Endourology
- Vol. 35 (9) , 1350-1356
Robot-Assisted Cystectomy and Ileal Conduit for Neurogenic Bladder: Comparison of Extracorporeal vs Intracorporeal Urinary Diversion
Objectives: The aim of the present study was to compare the perioperative outcomes of extracorporeal (EXTRA) vs intracorporeal (INTRA) urinary diversion in patients undergoing robotic cystectomy and ileal conduit for neurogenic bladder. Methods: All consecutive patients who underwent robot-assisted cystectomy and ileal conduit for neurogenic bladder in six centers between 2011 and 2020 were included in a retrospective study. Four centers performed exclusively INTRA urinary diversion all over the study period, one center performed exclusively EXTRA urinary diversion, and the last center performed EXTRA urinary diversion during a first period and INTRA during the subsequent period. Results: Ninety-seven patients were included: 66 in the EXTRA group and 31 in the INTRA group. There were 11 major (Clavien grade ≥3) postoperative complications in the overall population (11.3%) with no statistically significant difference between both groups (EXTRA = 12.1% vs INTRA = 9.7%; p = 0.99). The mean length of stay did not differ significantly between INTRA and EXTRA (13.1 vs 14.1 days; p = 0.44). The mean times to oral feeding and to return of bowel function were similar in the two groups (3.9 vs 3.5 days; p = 0.28 and 4.1 vs 4.1 days; p = 0.51, respectively). There was no incisional hernia in the INTRA group vs five in the EXTRA group (0% vs 7.6%; p = 0.17). Conclusion: The perioperative morbidity of robotic cystectomy and ileal conduit for neurogenic bladder appears to be relatively limited compared with the historical open series. Possibly due to the relatively small sample size, no difference was found between INTRA and EXTRA urinary diversion in terms of perioperative outcomes.
- 1 September 2021
- Published by SAGE Publications in Journal of Endourology
- Vol. 35 (9) , 1350-1356
Robot-Assisted Cystectomy and Ileal Conduit for Neurogenic Bladder: Comparison of Extracorporeal vs Intracorporeal Urinary Diversion
Objectives: The aim of the present study was to compare the perioperative outcomes of extracorporeal (EXTRA) vs intracorporeal (INTRA) urinary diversion in patients undergoing robotic cystectomy and ileal conduit for neurogenic bladder. Methods: All consecutive patients who underwent robot-assisted cystectomy and ileal conduit for neurogenic bladder in six centers between 2011 and 2020 were included in a retrospective study. Four centers performed exclusively INTRA urinary diversion all over the study period, one center performed exclusively EXTRA urinary diversion, and the last center performed EXTRA urinary diversion during a first period and INTRA during the subsequent period. Results: Ninety-seven patients were included: 66 in the EXTRA group and 31 in the INTRA group. There were 11 major (Clavien grade ≥3) postoperative complications in the overall population (11.3%) with no statistically significant difference between both groups (EXTRA = 12.1% vs INTRA = 9.7%; p = 0.99). The mean length of stay did not differ significantly between INTRA and EXTRA (13.1 vs 14.1 days; p = 0.44). The mean times to oral feeding and to return of bowel function were similar in the two groups (3.9 vs 3.5 days; p = 0.28 and 4.1 vs 4.1 days; p = 0.51, respectively). There was no incisional hernia in the INTRA group vs five in the EXTRA group (0% vs 7.6%; p = 0.17). Conclusion: The perioperative morbidity of robotic cystectomy and ileal conduit for neurogenic bladder appears to be relatively limited compared with the historical open series. Possibly due to the relatively small sample size, no difference was found between INTRA and EXTRA urinary diversion in terms of perioperative outcomes.
- Open Access
- 3 September 2020
- Published by International Scientific Information, Inc. in American Journal of Case Reports
- Vol. 21, e923416-1-e923416-4
Metal-Metal Hip Prosthesis and Kidney Cancer: Assumed Role of Chromium and Cobalt Overload
BACKGROUND:The case of a patient with bilateral renal cancers diagnosed at 94 and 120 months after metal-on-metal hip placement may serve as a warning. It suggests that there may be a need for kidney echography observation of patients with similar types of prostheses. CASE REPORT:A 61-year-old woman received a metal-on-metal hip prosthesis for degenerative arthritis in January 2007. In November 2014, after bleeding from the renal tract, she was diagnosed with clear cell carcinoma of the right kidney. When she returned to her orthopaedic surgeon 1 year later, a blood test showed a serum cobalt level that exceeded the French medical agency recommendation. After the patient’s metallic acetabulum was replaced in September 2015, her blood cobalt level fell. However, in February 2017, she was diagnosed with adenocarcinoma of the left kidney. Laser-induced breakdown spectroscopy (LIBS) showed cobalt, chromium, and silica overload in both the patient’s kidneys despite the drop in serum levels. CONCLUSIONS:In this case, exposure to a cobalt-chromium implant with high particulate wear, LIBS results showing chromium overload of the kidneys, diagnosis of renal cancer at 7 years, 10 months and 10 years in a patient with a metal-on-metal hip prosthesis suggests that there may be a causal relationship between the implant, carcinogenic chromium intoxication, and development of renal cancer.
- 3 September 2020
- Published by International Scientific Information, Inc. in American Journal of Case Reports
- Vol. 21, e923416-1-e923416-4
Open AccessMetal-Metal Hip Prosthesis and Kidney Cancer: Assumed Role of Chromium and Cobalt Overload
BACKGROUND:The case of a patient with bilateral renal cancers diagnosed at 94 and 120 months after metal-on-metal hip placement may serve as a warning. It suggests that there may be a need for kidney echography observation of patients with similar types of prostheses. CASE REPORT:A 61-year-old woman received a metal-on-metal hip prosthesis for degenerative arthritis in January 2007. In November 2014, after bleeding from the renal tract, she was diagnosed with clear cell carcinoma of the right kidney. When she returned to her orthopaedic surgeon 1 year later, a blood test showed a serum cobalt level that exceeded the French medical agency recommendation. After the patient’s metallic acetabulum was replaced in September 2015, her blood cobalt level fell. However, in February 2017, she was diagnosed with adenocarcinoma of the left kidney. Laser-induced breakdown spectroscopy (LIBS) showed cobalt, chromium, and silica overload in both the patient’s kidneys despite the drop in serum levels. CONCLUSIONS:In this case, exposure to a cobalt-chromium implant with high particulate wear, LIBS results showing chromium overload of the kidneys, diagnosis of renal cancer at 7 years, 10 months and 10 years in a patient with a metal-on-metal hip prosthesis suggests that there may be a causal relationship between the implant, carcinogenic chromium intoxication, and development of renal cancer.