Home

Main objective: develop a software system for computer aided and automatic diagnosis of the abdominal tumors based on multiple image modalities, involving both conventional and deep learning techniques.

Secondary objectives:

  • O1. Develop advanced image analysis and classification methods in order to attain maximum performance regarding the automatic diagnosis of some important abdominal malignant tumors, based on medical images.

  • O2. Compare the performances of the conventional and deep learning techniques in multiple situations, within various types of medical images of abdominal tumors.

  • O3. Diagnose the incipient tumors and the preneoplastic stages through appropriate methods.

  • O4. Sustain the research activity of the young researchers (young PhD students).


Workplan

The project activities are organized in three stages, corresponding to each year.

Stage 1 (May 12, 2022 - December 31, 2022).

  • The design of the software system for the automatic and computer aided diagnosis of the abdominal tumors. Preliminary experiments and results.

Stage 2 (January 1, 2023 - December 31, 2023).

  • The implementation of the software system for the automatic and computer aided diagnosis of the abdominal tumors.

Stage 3 (January 1, 2024 - May 10, 2024).

  • Installation of the software system at the beneficiary, its testing in real conditions and integration into clinical practice.


Publications

D. Mitrea, R. Brehar, R. Itu, S. Nedevschi, M. Socaciu, R. Badea, "Pancreatic Tumor Recognition from CT Images through Advanced Deep Learning Techniques", article accepted at IEEE International Conference on Automation, Quality and Testing, Robotics (AQTR 2024), Cluj-Napoca, May 16-18, 2024 (ISI Proceedings).

D. Mitrea, R. Brehar, S. Nedevschi, M. Platon-Lupsor, M. Socaciu, R. Badea, “Hepatocellular Carcinoma Recognition from Ultrasound Images Using Combinations of Conventional and Deep Learning Techniques", Sensors (ISI, Q2), Vol. 23, No. 5, pp. 1-29, https://www.mdpi.com/1424-8220/23/5/2520

D. Mitrea, R. Brehar, C. Mocan, S. Nedevschi, M. Socaciu, R. Badea, “Hepatocellular carcinoma recognition from ultrasound images by fusing convolutional neural networks at decision level”, The 46th International Conference on Telecommunications and Signal Processing (TSP 2023), 12-14 July 2023 (ISI Proceedings)

R. Brehar, D. Mitrea, S. Nedevschi, T. Moisoiu, F.I. Elec, M. Socaciu, “Kidney Tumor Segmentation and Grade Identification in CT Images”, 2023 IEEE 19th International Conference on Intelligent Computer Communication and Processing (ICCP), Cluj-Napoca, 26-28 Oct 2023 (ISI Proceedings)

D. Mitrea, V. Timu, V. A. Florian, C. Mocan, S. Nedevschi, M. Socaciu, R. Badea, “Liver tumor segmentation from Computed Tomography images through Convolutional Neural Networks”, The 2023 9th International Conference on Systems and Informatics (ICSAI 2023), 16-18 Dec 2023, Changsha, China, Proceedings, pp. 291-296, http://81.68.97.198/submission_icsai/ICSAI-2023-Proceeding.pdf.

D. Mitrea, S. Nedevschi, M. Socaciu, R. Badea, “Deep Learning Techniques for Liver Tumor Recognition in Ultrasound Images”, book chapter in the book titled “Deep Learning - Recent Findings and Researches”, IntechOpen (https://www.intechopen.com/online-first/deep-learning-techniques-for-liver-tumor-recognition-in-ultrasound-images)

D. Mitrea, R. Brehar, S. Nedevschi, M. Socaciu, R. Badea, "Hepatocellular Carcinoma recognition from ultrasound images through Convolutional Neural Networks and their combinations", International Conference on Advancements of Medicine and Health care through Technology, MediTech 2022, Cluj-Napoca, 20-22 Oct 2022, IFMBE Proceedings (IFMBE, volume 102), pp. 3-11, Springer, https://link.springer.com/chapter/10.1007/978-3-031-51120-2_1

R. Brehar, D. Mitrea, S. Nedevschi, M. Socaciu, T. Moisoiu, F. Elec, "Kidney Tumor Stage Identification in CT Images", IEEE Transactions on Medical Imaging (under assessment)


European Patent:

D. Mitrea, R. Brehar, R. Itu, A.-V. Florian, M. Socaciu, T. Moisoiu, “Improving the Performance of Abdominal Tumors Diagnosis within Medical Images through the Combination of Conventional and Deep Learning Techniques”, proposal submitted to European Patent Office (EPO)

Results and Conclusions

In the context of the ACADTUM research project, the research team developed and experimented advanced computerized methods for the automated and computer aided diagnosis of the abdominal tumors, based on medical images of various types: ultrasound (US), computer tomography (CT), magnetic resonance images (MRI). Aiming to perform abdominal tumor recognition and segmentation within medical images, representative Convolutional Neural Networks (CNN) based techniques, original CNN architectures, as well as CNN combinations, at classifier and decision level, were considered for this purpose. Original conventional techniques, based on advanced texture analysis methods, were experimented as well, being compared, respectively combined with the deep-learning methods. Important steps have been performed to automatically detect the renal tumors’ evolution stages, respectively the pre-neoplastic states, in the case of liver cancer. The best performing techniques were integrated within the ACADTUM software system, destined for the automatic and computer aided diagnosis of abdominal tumors. The automatic recognition and segmentation methods led to a superior performance when being assessed on CT and MRI images. However, the value of the ultrasound imaging based medical examination techniques should not be ignored, being known that ultrasonography represents a non-invasive, low-cost, safe medical investigation method, suitable for disease evolution monitoring.


Contact

Delia Mitrea

Address:
Technical University of Cluj-Napoca
Computer Science department,
Memorandumului st., no. 28, 400 114, Cluj-Napoca
Romania

Office: Baritiu st., no. 26

E-mail:
Delia.Mitrea@cs.utcluj.ro



Full project name:
ACADTUM - Automatic and computer aided diagnosis of abdominal tumors, through advanced machine learning, within various types of medical images - PN-III-P1-1.1-TE-2021-1293

Contract:
TE 156/2022

Funding:
Ministry of Research and Innovation, CNCS – UEFISCDI