Pressure injury image analysis with machine learning techniques: A systematic review on previous and possible future methods

Zahia, Sofia; Zapirain, Maria Begoña Garcia; Sevillano, Xavier; González, Alejandro; Kim, Paul J.; Elmaghraby, Adel

doi:10.1016/j.artmed.2019.101742

Cited by 58 publications

(40 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, deep learning models [7,[14][15][16][17] have made remarkable progress in computer vision, specifically in biomedical image processing, due to their abilities to automatically learn complicated and advanced features from images, which inspired various researchers to leverage these models in the classification of breast cancer histopathology images [7]. Especially convolutional neural networks (CNNs) [18] are widely used in image-related tasks due to their abilities to effectively share parameters across various layers within a deep learning model.…”

Section: Introductionmentioning

confidence: 99%

Breast Cancer Histopathology Image Classification Using an Ensemble of Deep Learning Models

Hameed

Zahia

Garcia‐Zapirain

et al. 2020

Sensors

Self Cite

184

View full text Add to dashboard Cite

Breast cancer is one of the major public health issues and is considered a leading cause of cancer-related deaths among women worldwide. Its early diagnosis can effectively help in increasing the chances of survival rate. To this end, biopsy is usually followed as a gold standard approach in which tissues are collected for microscopic analysis. However, the histopathological analysis of breast cancer is non-trivial, labor-intensive, and may lead to a high degree of disagreement among pathologists. Therefore, an automatic diagnostic system could assist pathologists to improve the effectiveness of diagnostic processes. This paper presents an ensemble deep learning approach for the definite classification of non-carcinoma and carcinoma breast cancer histopathology images using our collected dataset. We trained four different models based on pre-trained VGG16 and VGG19 architectures. Initially, we followed 5-fold cross-validation operations on all the individual models, namely, fully-trained VGG16, fine-tuned VGG16, fully-trained VGG19, and fine-tuned VGG19 models. Then, we followed an ensemble strategy by taking the average of predicted probabilities and found that the ensemble of fine-tuned VGG16 and fine-tuned VGG19 performed competitive classification performance, especially on the carcinoma class. The ensemble of fine-tuned VGG16 and VGG19 models offered sensitivity of 97.73% for carcinoma class and overall accuracy of 95.29%. Also, it offered an F1 score of 95.29%. These experimental results demonstrated that our proposed deep learning approach is effective for the automatic classification of complex-natured histopathology images of breast cancer, more specifically for carcinoma images.

show abstract

Section: Introductionmentioning

confidence: 99%

Breast Cancer Histopathology Image Classification Using an Ensemble of Deep Learning Models

Hameed

Zahia

Garcia‐Zapirain

et al. 2020

Sensors

Self Cite

184

View full text Add to dashboard Cite

show abstract

“…Traditionally, wound image segmentation was performed using image processing techniques [ 3 ]. Several image processing approaches have been proposed in the literature, such as Region-based segmentation [ 14 , 15 , 16 , 17 ], and edge-based segmentation [ 14 , 18 , 19 ].…”

Section: Related Workmentioning

confidence: 99%

“…Pressure injuries represent one of the greatest challenges in the healthcare sector, as they are considered one of the major causes of death, and a financial burden on healthcare systems [ 1 , 2 ]. They are chronic wounds resulting from damage caused by pressure over time causing an ischemia of underlying skin structures, especially on bony areas [ 3 ], as shown in Figure 1 . Pressure injury stages varies from stage 1 to stage 4 depending on the deepest parts of the ulcer and the type of tissue affected.…”

Section: Introductionmentioning

confidence: 99%

Integrating 3D Model Representation for an Accurate Non-Invasive Assessment of Pressure Injuries with Deep Learning

Zahia

Garcia‐Zapirain

Elmaghraby

2020

Sensors

Self Cite

View full text Add to dashboard Cite

Pressure injuries represent a major concern in many nations. These wounds result from prolonged pressure on the skin, which mainly occur among elderly and disabled patients. If retrieving quantitative information using invasive methods is the most used method, it causes significant pain and discomfort to the patients and may also increase the risk of infections. Hence, developing non-intrusive methods for the assessment of pressure injuries would represent a highly useful tool for caregivers and a relief for patients. Traditional methods rely on findings retrieved solely from 2D images. Thus, bypassing the 3D information deriving from the deep and irregular shape of this type of wounds leads to biased measurements. In this paper, we propose an end-to-end system which uses a single 2D image and a 3D mesh of the pressure injury, acquired using the Structure Sensor, and outputs all the necessary findings such as: external segmentation of the wound as well as its real-world measurements (depth, area, volume, major axis and minor axis). More specifically, a first block composed of a Mask RCNN model uses the 2D image to output the segmentation of the external boundaries of the wound. Then, a second block matches the 2D and 3D views to segment the wound in the 3D mesh using the segmentation output and generates the aforementioned real-world measurements. Experimental results showed that the proposed framework can not only output refined segmentation with 87% precision, but also retrieves reliable measurements, which can be used for medical assessment and healing evaluation of pressure injuries.

show abstract

“…Since there is a dramatic variations in the technology since last decades, medical images are also gathered along with medical data records. Authors of [7] and [8] focuses more on medical images and use machine learning algorithms for their proposals. As mentioned in [7], ML in artificial intelligence proved to be one of the promising techniques to provide selfinterpretation of images, quality extraction of data and predictions for a patient related to previous medical records.…”

Section: Introductionmentioning

confidence: 99%

“…As mentioned in [7], ML in artificial intelligence proved to be one of the promising techniques to provide selfinterpretation of images, quality extraction of data and predictions for a patient related to previous medical records. Here medical records may be in the form of images [8] since by using deep learning system, it is accurate to make assessments regarding either approaches towards a system in which predictions are based on the processing of images (for example injury images),] or recommending some medical facilities to them.…”

Section: Introductionmentioning

confidence: 99%

Performance Analysis of Supervised Machine Learning Algorithms on Medical Dataset

Juyal*,

Pandey,

Pant

et al. 2020

IJRTE

View full text Add to dashboard Cite

Machine learning (ML) algorithms are designed to perform prediction based on features. With the help of machine learning, system can automatically learn and improve by experience. Machine learning comes under Artificial intelligence. Machine learning is broadly categorized in two types: supervised and unsupervised. Supervised ML performs classification and unsupervised is for clustering. In present scenario, machine learning is used in various areas. It can be used for biometric recognition, hand writing recognition, medical diagnosis etc. In medical field, machine learning plays an important role in identifying diseases based on patient’s features. Presently,doctors use software application based on machine learning algorithm in various disease diagnosis like cancer, cardiac arrest and many more. In this paper we used an ensemble learning method to predict heart problem. Our study described the performance of ML algorithms by comparing various evaluating parameters such as F-measure, Recall, ROC, precision and accuracy. The study done with various combination ML classifiers such as, Decision Tree (DT), Naïve Bayes (NB), Support Vector Machine (SVM), Random Forest (RF) algorithm to predict heart problem. The result showed that by combining two ML algorithm, DT with NB, 81.1% accuracy was achieved. Simultaneously, the models like Support Vector machine (SVM), Decision tree, Naïve Bayes, Random Forest models were also trained and tested individually.

show abstract

Pressure injury image analysis with machine learning techniques: A systematic review on previous and possible future methods

Cited by 58 publications

References 43 publications

Breast Cancer Histopathology Image Classification Using an Ensemble of Deep Learning Models

Breast Cancer Histopathology Image Classification Using an Ensemble of Deep Learning Models

Integrating 3D Model Representation for an Accurate Non-Invasive Assessment of Pressure Injuries with Deep Learning

Performance Analysis of Supervised Machine Learning Algorithms on Medical Dataset

Contact Info

Product

Resources

About