Boosting CNN Learning by Ensemble Image Preprocessing Methods for Cervical Cancer Segmentation

Bnouni, Nesrine; Amor, Hadil Ben; Rekik, Islem; Rhim, Mohamed Salah; Solaiman, Basel; Amara, Najoua Essoukri Ben

doi:10.1109/ssd52085.2021.9429422

Cited by 11 publications

(8 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typically, the final feature map output ends up being 32 times smaller in each spatial dimension than the original image. Figure 1b–d demonstrates the encoder–decoder architectures of fully convolutional networks (FCN), U‐Net, and context encoder network (CE‐Net) for comparison 2 4–26 …”

Section: Methodsmentioning

confidence: 99%

“…With the development and wide application of deep learning, deep learning–based automatic segmentation has shown a superior performance in the reduction of target volume delineation variation for many tumors 14–16 . As for cervical cancer, three paralleled convolutional neural networks (CNNs) with the same architecture trained following different image preprocessing methods had been applied 17,18 . However, CNNs suffer from the problem of reducing the resolution of original images while increasing the ambiguity of object boundaries inevitably 19 .…”

Section: Introductionmentioning

confidence: 99%

“…[14][15][16] As for cervical cancer, three paralleled convolutional neural networks (CNNs) with the same architecture trained following different image preprocessing methods had been applied. 17,18 However, CNNs suffer from the problem of reducing the resolution of original images while increasing the ambiguity of object boundaries inevitably. 19 Recently, the lightweight RefineNet was introduced to refine object detectors for autonomous driving, which generates high-resolution semantic feature by fusing coarse high-level features with finer grained low-level features.…”

mentioning

confidence: 99%

See 2 more Smart Citations

RefineNet‐based 2D and 3D automatic segmentations for clinical target volume and organs at risks for patients with cervical cancer in postoperative radiotherapy

Xiao

Jin

et al. 2022

J Applied Clin Med Phys

View full text Add to dashboard Cite

Purpose An accurate and reliable target volume delineation is critical for the safe and successful radiotherapy. The purpose of this study is to develop new 2D and 3D automatic segmentation models based on RefineNet for clinical target volume (CTV) and organs at risk (OARs) for postoperative cervical cancer based on computed tomography (CT) images. Methods A 2D RefineNet and 3D RefineNetPlus3D were adapted and built to automatically segment CTVs and OARs on a total of 44 222 CT slices of 313 patients with stage I–III cervical cancer. Fully convolutional networks (FCNs), U‐Net, context encoder network (CE‐Net), UNet3D, and ResUNet3D were also trained and tested with randomly divided training and validation sets, respectively. The performances of these automatic segmentation models were evaluated by Dice similarity coefficient (DSC), Jaccard similarity coefficient, and average symmetric surface distance when comparing them with manual segmentations with the test data. Results The DSC for RefineNet, FCN, U‐Net, CE‐Net, UNet3D, ResUNet3D, and RefineNet3D were 0.82, 0.80, 0.82, 0.81, 0.80, 0.81, and 0.82 with a mean contouring time of 3.2, 3.4, 8.2, 3.9, 9.8, 11.4, and 6.4 s, respectively. The generated RefineNetPlus3D demonstrated a good performance in the automatic segmentation of bladder, small intestine, rectum, right and left femoral heads with a DSC of 0.97, 0.95, 091, 0.98, and 0.98, respectively, with a mean computation time of 6.6 s. Conclusions The newly adapted RefineNet and developed RefineNetPlus3D were promising automatic segmentation models with accurate and clinically acceptable CTV and OARs for cervical cancer patients in postoperative radiotherapy.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

RefineNet‐based 2D and 3D automatic segmentations for clinical target volume and organs at risks for patients with cervical cancer in postoperative radiotherapy

Xiao

Jin

et al. 2022

J Applied Clin Med Phys

View full text Add to dashboard Cite

show abstract

“…For a computer-aided analysis of EC detection a practical technique is to be implemented to use a CNN classifier for EC. Bnouni et al [44] proposed that CNN can be boosted by ensemble image pre-processing methods and CC can be segmented. The proposed boosting strategy is a fully automated novel ensemble pre-processing for CC segmentation.…”

Section: Existing Methodsmentioning

confidence: 99%

Image processing and machine learning approaches for the automatic diagnosis of endometrial cancer

Libish²

2022

ijhs

View full text Add to dashboard Cite

In many medical imaging based diagnosis, Deep Learning (DL) algorithms play an essential role. A DL algorithm can be used to identify abnormal and normal cells in the uterus's endometrium in order to discover Endometrial Cancer (EC) cells. EC is difficult to diagnose since it develops without causing any symptoms. DL algorithm can distinguish between normal, abnormal, and malignant cells, producing more accurate findings than screening by hand procedures such as liquid cytology and Pap smear test. For the accurate and easier detection of EC cells, DL employs multiple architectures. The findings of an analysis and survey of the many forms of DL architecture, as well as their accuracy and performance, are addressed in this work. Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) are the most modalities of advanced imaging used for the non-invasive diagnosis of EC. EC is considered as the fourth most prevalent malignancy in women and one of the common most gynaecological cancers. Diagnostic imaging in clinical evaluation with has not been proven yet to be precise enough to substitute surgical staging in determining the spread of cancer. It may allow for improved surgical process optimization and a more customised therapeutic plan.

show abstract

“…A common cytological auxiliary screening method for cervical cancer is to promote the detection and grading of cervical cancer using graph-based methods based on the segmentation results of complex non-convex regions [49]. Bnouni et al [50] proposed a collection preconditioning method to realize the segmentation of cervical cancer cells based on a CNN. Subsequent scholars have continued this idea, and Sellamuthu et al [51] proposed an improved deep learning algorithm based on a double-tree complex wavelet transform (DTCWT).…”

Section: Segmentation Of Pathological Cellsmentioning

confidence: 99%

High Precision Cervical Precancerous Lesion Classification Method Based on ConvNeXt

Tang,

Zhang,

Gong

et al. 2023

Bioengineering

View full text Add to dashboard Cite

Traditional cervical cancer diagnosis mainly relies on human papillomavirus (HPV) concentration testing. Considering that HPV concentrations vary from individual to individual and fluctuate over time, this method requires multiple tests, leading to high costs. Recently, some scholars have focused on the method of cervical cytology for diagnosis. However, cervical cancer cells have complex textural characteristics and small differences between different cell subtypes, which brings great challenges for high-precision screening of cervical cancer. In this paper, we propose a high-precision cervical cancer precancerous lesion screening classification method based on ConvNeXt, utilizing self-supervised data augmentation and ensemble learning strategies to achieve cervical cancer cell feature extraction and inter-class discrimination, respectively. We used the Deep Cervical Cytological Levels (DCCL) dataset, which includes 1167 cervical cytology specimens from participants aged 32 to 67, for algorithm training and validation. We tested our method on the DCCL dataset, and the final classification accuracy was 8.85% higher than that of previous advanced models, which means that our method has significant advantages compared to other advanced methods.

show abstract

Boosting CNN Learning by Ensemble Image Preprocessing Methods for Cervical Cancer Segmentation

Cited by 11 publications

References 25 publications

RefineNet‐based 2D and 3D automatic segmentations for clinical target volume and organs at risks for patients with cervical cancer in postoperative radiotherapy

RefineNet‐based 2D and 3D automatic segmentations for clinical target volume and organs at risks for patients with cervical cancer in postoperative radiotherapy

Image processing and machine learning approaches for the automatic diagnosis of endometrial cancer

High Precision Cervical Precancerous Lesion Classification Method Based on ConvNeXt

Contact Info

Product

Resources

About