A Multiple Instance Learning Algorithm for Color Images Classification

Astorino, Annabella; Fuduli, Antonio; Gaudioso, Manlio; Vocaturo, Eugenio

doi:10.1145/3216122.3216144

Cited by 26 publications

(10 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although in [34] it has been shown that using some pre-processing methodologies on the images improves the classification performances, in our experiments we preferred do not use such techniques. In this way it was intended to operate in the worst conditions in order to compare the classification performances obtainable by using, on one hand, only color features (as proposed in [40,41,43]) and, on the other hand, color and texture features. The fact that the used data set consists of plain photographs, instead of high-quality dermatoscopic images such as those ones constituting the Ph 2 database, justifies the obtained classification performance.…”

Section: Discussionmentioning

confidence: 99%

“…Such a method belongs the class of the Multiple Instance Learning (MIL) algorithms. The MIL paradigm (see for example [38,39]) is a relatively new approach for classification problems which fits very well on image classification, as shown in the recent works [40,41] and, in particular, also for medical images and video analysis [42]. It differs from the classical supervised classification approaches, since it is aimed at classifying sets of items: such sets are called bags and the items inside the sets are called instances.…”

Section: Numerical Experimentationsmentioning

confidence: 99%

See 1 more Smart Citation

Melanoma detection using color and texture features in computer vision systems

Fuduli¹,

Veltri²,

Vocaturo³

et al. 2019

Adv. sci. technol. eng. syst. j.

Self Cite

View full text Add to dashboard Cite

All forms of skin cancer are becoming widespread. These forms of cancer, and melanoma in particular, are insidious and aggressive and if not treated promptly can be lethal to humans. Effective treatment of skin lesions depends strongly on the timeliness of the diagnosis: for this reason, artificial vision systems are required to play a crucial role in supporting the diagnosis of skin lesions. This work offers insights into the state of the art in the field of melanoma image classification. We include a numerical section where a preliminary analysis of some classification techniques is performed, using color and texture features on a data set constituted by plain photographies, to which no pre-processing technique has been applied. This is motivated by the necessity to open new horizons in creating self-diagnosis systems for accessible skin lesions, due also to a huge innovation of cameras, smartphones technology and wearable devices.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Numerical Experimentationsmentioning

confidence: 99%

Melanoma detection using color and texture features in computer vision systems

Fuduli¹,

Veltri²,

Vocaturo³

et al. 2019

Adv. sci. technol. eng. syst. j.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Random forest constructs a multitude of decision trees at training and output the class the individual trees [42]. Besides, Multiple Instance Learning (MIL) has been widely used in medical image classification [43]- [45], Gaudioso et alproven that multiple instance learning can helps to classify data belonging to similar categories [45]. Multi-Layer perceptron (MLP) [46] or neural network (NN) consists of multiple units neurons, arranged in layers, and trained by the backpropagation algorithm [47].…”

Section: B Classification Algorithms In Machine Learningmentioning

confidence: 99%

Eliminating Indefiniteness of Clinical Spectrum for Better Screening COVID-19

Guo

Liu

Zhao

et al. 2021

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

The coronavirus disease 2019 (COVID-19) has swept all over the world in the last few months. Due to the limited detection facilities and medical resources, especially in developing countries, a large number of suspected cases can only receive common clinical diagnosis rather than more effective detections like Reverse Transcription Polymerase Chain Reaction (RT-PCR) tests or CT scans. This motivates us to develop a quick screening method on suspected patients via common clinical diagnosis results. However, the diagnostic items of different patients may vary greatly and there is a huge variation in the dimension of the diagnosis data among different suspected patients, which makes it hard to process these indefinite dimension data via classical classification algorithms. To resolve this problem, we propose an Indefiniteness Elimination Network (IE-Net) to eliminate the influence of the varied dimensions and make predictions about the COVID-19 cases. The IE-Net is in an encoder-decoder framework fashion and an indefiniteness elimination operation is proposed to transfer the indefinite dimension feature into a fixed dimension feature. Comprehensive experiments were conducted on the public available COVID-19 Clinical Spectrum dataset. Experimental results show that the proposed indefiniteness elimination operation greatly improves the classification performance and the IE-Net achieves 94.80% accuracy, 92.79% recall, 92.97% precision and 94.93% AUC for distinguishing COVID-19 cases from non-COVID-19 cases with only common clinical diagnose data. We further compared our methods with 3 classical classification algorithms: random forest, gradient boosting and multi-layer perceptron (MLP). Compared with traditional classification methods, the proposed IE-Net can bring at least 9.40% accuracy gains, 12.17% recall gains, 3.66% precision and 9.71% AUC gains. To explore the specificity of each clinical test item, we further analyzed the possible relationship between each clinical test item and COVID-19. In general, we proposed a novel image complementary method for COVID-19 screen-

show abstract

“…The MIL paradigm finds application in a lot of fields: text categorization, image recognition (Astorino et al 2017(Astorino et al , 2018, video analysis, diagnostics by means of images (Astorino et al 2019b, and Quellec et al 2017) and so on. An example fitting very well the standard MIL assumption stated above is in discriminating between healthy and nonhealthy patients on the basis of their medical scan (bag): if at least a region (instance) of the medical scan (bag) is abnormal, then the patient is classified as nonhealthy and, on the contrary, when all the regions (instances) of the medical scan (bag) are normal, then the patient is classified as healthy.…”

Section: Introductionmentioning

confidence: 99%

A heuristic approach for multiple instance learning by linear separation

et al. 2022

Self Cite

View full text Add to dashboard Cite

We present a fast heuristic approach for solving a binary multiple instance learning (MIL) problem, which consists in discriminating between two kinds of item sets: the sets are called bags and the items inside them are called instances. Assuming that only two classes of instances are allowed, a common standard hypothesis states that a bag is positive if it contains at least a positive instance and it is negative when all its instances are negative. Our approach constructs a MIL separating hyperplane by preliminary fixing the normal and reducing the learning phase to a univariate nonsmooth optimization problem, which can be quickly solved by simply exploring the kink points. Numerical results are presented on a set of test problems drawn from the literature.

show abstract

A Multiple Instance Learning Algorithm for Color Images Classification

Cited by 26 publications

References 22 publications

Melanoma detection using color and texture features in computer vision systems

Melanoma detection using color and texture features in computer vision systems

Eliminating Indefiniteness of Clinical Spectrum for Better Screening COVID-19

A heuristic approach for multiple instance learning by linear separation

Contact Info

Product

Resources

About