Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations Generators
Mathematical modelling to compute ground truth from 3D images is an area of research that can strongly benefit from machine learning methods. Deep neural networks (DNNs) are state-of-the-art methods design for solving these kinds of difficulties. Convolutional neural networks (CNNs), as one class of...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-03-01
|
| Series: | Symmetry |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-8994/12/3/416 |
| _version_ | 1818001080710070272 |
|---|---|
| author | Omar Bilalovic Zikrija Avdagic Samir Omanovic Ingmar Besic Vedad Letic Christophe Tatout |
| author_facet | Omar Bilalovic Zikrija Avdagic Samir Omanovic Ingmar Besic Vedad Letic Christophe Tatout |
| author_sort | Omar Bilalovic |
| collection | DOAJ |
| description | Mathematical modelling to compute ground truth from 3D images is an area of research that can strongly benefit from machine learning methods. Deep neural networks (DNNs) are state-of-the-art methods design for solving these kinds of difficulties. Convolutional neural networks (CNNs), as one class of DNNs, can overcome special requirements of quantitative analysis especially when image segmentation is needed. This article presents a system that uses a cascade of CNNs with symmetric blocks of layers in chain, dedicated to 3D image segmentation from microscopic images of 3D nuclei. The system is designed through eight experiments that differ in following aspects: number of training slices and 3D samples for training, usage of pre-trained CNNs and number of slices and 3D samples for validation. CNNs parameters are optimized using linear, brute force, and random combinatorics, followed by voter and median operations. Data augmentation techniques such as reflection, translation and rotation are used in order to produce sufficient training set for CNNs. Optimal CNN parameters are reached by defining 11 standard and two proposed metrics. Finally, benchmarking demonstrates that CNNs improve segmentation accuracy, reliability and increased annotation accuracy, confirming the relevance of CNNs to generate high-throughput mathematical ground truth 3D images. |
| first_indexed | 2024-04-14T03:30:07Z |
| format | Article |
| id | doaj.art-687e8e5afe91466984376c74217d81ac |
| institution | Directory Open Access Journal |
| issn | 2073-8994 |
| language | English |
| last_indexed | 2024-04-14T03:30:07Z |
| publishDate | 2020-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Symmetry |
| spelling | doaj.art-687e8e5afe91466984376c74217d81ac2022-12-22T02:15:00ZengMDPI AGSymmetry2073-89942020-03-0112341610.3390/sym12030416sym12030416Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations GeneratorsOmar Bilalovic0Zikrija Avdagic1Samir Omanovic2Ingmar Besic3Vedad Letic4Christophe Tatout5Computer Science, Faculty of Electrical Engineering, University of Sarajevo, 71000 Sarajevo, Bosnia and HerzegovinaComputer Science, Faculty of Electrical Engineering, University of Sarajevo, 71000 Sarajevo, Bosnia and HerzegovinaComputer Science, Faculty of Electrical Engineering, University of Sarajevo, 71000 Sarajevo, Bosnia and HerzegovinaComputer Science, Faculty of Electrical Engineering, University of Sarajevo, 71000 Sarajevo, Bosnia and HerzegovinaComputer Science, Faculty of Electrical Engineering, University of Sarajevo, 71000 Sarajevo, Bosnia and HerzegovinaUniversité Clermont Auvergne, CNRS, Inserm, GReD, F-63000 Clermont-Ferrand, FranceMathematical modelling to compute ground truth from 3D images is an area of research that can strongly benefit from machine learning methods. Deep neural networks (DNNs) are state-of-the-art methods design for solving these kinds of difficulties. Convolutional neural networks (CNNs), as one class of DNNs, can overcome special requirements of quantitative analysis especially when image segmentation is needed. This article presents a system that uses a cascade of CNNs with symmetric blocks of layers in chain, dedicated to 3D image segmentation from microscopic images of 3D nuclei. The system is designed through eight experiments that differ in following aspects: number of training slices and 3D samples for training, usage of pre-trained CNNs and number of slices and 3D samples for validation. CNNs parameters are optimized using linear, brute force, and random combinatorics, followed by voter and median operations. Data augmentation techniques such as reflection, translation and rotation are used in order to produce sufficient training set for CNNs. Optimal CNN parameters are reached by defining 11 standard and two proposed metrics. Finally, benchmarking demonstrates that CNNs improve segmentation accuracy, reliability and increased annotation accuracy, confirming the relevance of CNNs to generate high-throughput mathematical ground truth 3D images.https://www.mdpi.com/2073-8994/12/3/416mathematical ground truth imagedeep learningconvolutional neural networkmicroscopic imagesoptimization |
| spellingShingle | Omar Bilalovic Zikrija Avdagic Samir Omanovic Ingmar Besic Vedad Letic Christophe Tatout Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations Generators Symmetry mathematical ground truth image deep learning convolutional neural network microscopic images optimization |
| title | Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations Generators |
| title_full | Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations Generators |
| title_fullStr | Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations Generators |
| title_full_unstemmed | Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations Generators |
| title_short | Mathematical Modelling of Ground Truth Image for 3D Microscopic Objects Using Cascade of Convolutional Neural Networks Optimized with Parameters' Combinations Generators |
| title_sort | mathematical modelling of ground truth image for 3d microscopic objects using cascade of convolutional neural networks optimized with parameters combinations generators |
| topic | mathematical ground truth image deep learning convolutional neural network microscopic images optimization |
| url | https://www.mdpi.com/2073-8994/12/3/416 |
| work_keys_str_mv | AT omarbilalovic mathematicalmodellingofgroundtruthimagefor3dmicroscopicobjectsusingcascadeofconvolutionalneuralnetworksoptimizedwithparameterscombinationsgenerators AT zikrijaavdagic mathematicalmodellingofgroundtruthimagefor3dmicroscopicobjectsusingcascadeofconvolutionalneuralnetworksoptimizedwithparameterscombinationsgenerators AT samiromanovic mathematicalmodellingofgroundtruthimagefor3dmicroscopicobjectsusingcascadeofconvolutionalneuralnetworksoptimizedwithparameterscombinationsgenerators AT ingmarbesic mathematicalmodellingofgroundtruthimagefor3dmicroscopicobjectsusingcascadeofconvolutionalneuralnetworksoptimizedwithparameterscombinationsgenerators AT vedadletic mathematicalmodellingofgroundtruthimagefor3dmicroscopicobjectsusingcascadeofconvolutionalneuralnetworksoptimizedwithparameterscombinationsgenerators AT christophetatout mathematicalmodellingofgroundtruthimagefor3dmicroscopicobjectsusingcascadeofconvolutionalneuralnetworksoptimizedwithparameterscombinationsgenerators |