Hospital Pedro Hispano Research Articles

A strong start: The benefits of coping strategies for neonatal intensive care unit families

Health, relationships and bonding between parents and infants are at risk when preterm babies need intensive care for extended periods, affecting child’s development and parental mental health. Pedro Hispano Hospital Neonatal Team trough “VIP neo project”, uses bonding hearts and NICU diaries to address these issues. Parents of newborns admitted to NICU between May and October 2022 were included. They reported enhanced self-assurance and reduced anxiety (94.1%) using NICU diaries and 17.2% reported use by other family members. Similarly, fostered connection (83.3%) and reduced anxiety (75%) was described

Hyperbaric oxygen therapy in malignant otitis externa: a retrospective analysis.

Malignant otitis externa (MOE) is a rare form of invasive osteomyelitis of the external ear canal. It is typically caused by Pseudomonas aeruginosa in immunocompromised patients. The diagnosis is clinical, and the initial treatment involves systemic antibiotics or antifungal therapy. Surgery is usually only considered when medical treatment has failed. Although hyperbaric oxygen therapy (HBOT) is recommended for refractory osteomyelitis, there are no specific guidelines for MOE. This is a retrospective study that evaluates clinical data, treatment, and results obtained in patients diagnosed with MOE treated with HBOT at the Pedro Hispano Hospital between 2007 and 2022. During the study period, fifteen patients diagnosed with MOE were admitted for treatment with HBOT. All patients received antibiotic and/or antifungal therapy, and three required surgical intervention before starting HBOT. The pathology was successfully managed on all patients. HBOT may be an effective adjuvant treatment option in patients with MOE but it lacks robust scientific evidence. However, its therapeutic value should not be underestimated due to the good results and few adverse effects reported in recent retrospective studies and case reports.

Effects of the COVID-19 Pandemic on Breastfeeding Initiation and Duration: A Retrospective Cohort Study.

Several studies support the health benefits of breastfeeding for both the mother and the newborn. However, a significant number of mothers discontinue breastfeeding within the first six months of childbirth, with several factors influencing breastfeeding adherence.The purpose of this study is to assess the impact of the COVID-19 pandemic on the prevention of mother-to-newborn infection transmission, breastfeeding patterns and duration, and the incidence of other infections during the first year of life. Data from a sample of 39 mothers who gave birth at the Hospital Pedro Hispano in Porto, Portugual, between March 2020 and November 2021 were collected and a telephone questionnaire was administered. Statistical analysis was conducted using R software, v. 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria). In terms of the impact of the COVID-19 norm 18/2020, which went into effect on March 30th, our research found that the type of feeding during hospitalization was significantly influenced by this norm (X2=10.30, p=0.006). We also confirmed that mothers who received home assistance breastfed for an extra 4.5 months (95% CI: 1-7.5) compared with mothers who did not receive such assistance. Regarding the effect of COVID-19 and breastfeeding on newborn health, our study found that if the total duration of breastfeeding is less than six months, an infection is approximately five times more likely (95% CI = 1.06- 29.56). Overall, the findings of this study indicate that the efforts implemented at Hospital Pedro Hispano to limit the effects of the COVID-19 pandemic had some effect on immediate breastfeeding patterns, but not on the total duration of breastfeeding or newborn health. Nonetheless, more continuous assistance at home would have been beneficial.

Skin lesion classification using HG-PSO and YOLOv7 based convolutional network in real time.

Skin cancer is a chronic illness seen visually and further diagnosed with a dermoscopic examination. It is crucial to precisely localize and classify lesions from dermoscopic images to diagnose and treat skin cancers as soon as possible. This work presents melanoma identification, and the classification method significantly improves accuracy and precision. This work proposes a method Hybrid of Genetic and Particle swarm optimization (HG-PSO), and You only look once version 7 (YOLOv7) based convolutional network for skin cancer classification. The infected region is first located using optimized YOLOv7 object detection. Then color thresholding is applied to segment it, which is passed to the proposed convolutional network for classification. This work is tested on the Human Against Machine with 10,000 training images (HAM10000), International Skin Imaging Collaboration (ISIC)-2019, and Hospital Pedro Hispano (PH2) datasets, and the findings are compared to the state-of-the-art methods for classifying skin cancer. The proposed method achieves 98.86% accuracy, 99.00% average precision, 98.85% average recall, and 98.85% average F1-score on the HAM10000 dataset. It achieves 97.10% accuracy on ISIC-2019 datasets. The average precision obtained is 97.37%, the average recall is 97.13%, and the average F1-score is 97.13% on the ISIC-2019 dataset. It achieves a 97.7% accuracy on the PH2 dataset. The average precision obtained is 99.00%, the average recall is 96.00%, and the average F1-score is 97.00% on the PH2 dataset. The test time taken by this method on datasets HAM10000, ISIC-2019, and PH2 dataset is 2, 3, and 2 s, respectively, which may help give faster responses in telemedicine.

Study protocol for a pilot randomised controlled trial evaluating the feasibility and effectiveness of non-pharmacological interventions to recover functionality after a transient ischaemic attack or a minor stroke: the ‘Back to Normal’ trial

IntroductionTransient ischaemic attack (TIA) and minor stroke are frequently assumed as temporary or non-disabling events. However, evidence suggests that these patients can experience relevant impairment and functional disability. Therefore, the...

A Hybrid Stacked Restricted Boltzmann Machine with Sobel Directional Patterns for Melanoma Prediction in Colored Skin Images

Melanoma, a kind of skin cancer that is very risky, is distinguished by uncontrolled cell multiplication. Melanoma detection is of the utmost significance in clinical practice because of the atypical border structure and the numerous types of tissue it can involve. The identification of melanoma is still a challenging process for color images, despite the fact that numerous approaches have been proposed in the research that has been done. In this research, we present a comprehensive system for the efficient and precise classification of skin lesions. The framework includes preprocessing, segmentation, feature extraction, and classification modules. Preprocessing with DullRazor eliminates skin-imaging hair artifacts. Next, Fully Connected Neural Network (FCNN) semantic segmentation extracts precise and obvious Regions of Interest (ROIs). We then extract relevant skin image features from ROIs using an enhanced Sobel Directional Pattern (SDP). For skin image analysis, Sobel Directional Pattern outperforms ABCD. Finally, a stacked Restricted Boltzmann Machine (RBM) classifies skin ROIs. Stacked RBMs accurately classify skin melanoma. The experiments have been conducted on five datasets: Pedro Hispano Hospital (PH2), International Skin Imaging Collaboration (ISIC 2016), ISIC 2017, Dermnet, and DermIS, and achieved an accuracy of 99.8%, 96.5%, 95.5%, 87.9%, and 97.6%, respectively. The results show that a stack of Restricted Boltzmann Machines is superior for categorizing skin cancer types using the proposed innovative SDP.

1362. Lyme Disease and Neurologic Manifestations in a District Hospital in Portugal

Abstract Background Lyme disease (LD), caused by Borrelia burgdorferi sensu lato’s species, has a wide presentation and severity and is often a neglected diagnosis. Neuroborreliosis (NB) occurs in 10-15% of patients (pts). This study evaluate NB (clinical manifestations, treatment and evolution) in a district acute hospital in Portugal. Methods Pts with Borrelia burgdorferi antibodies detected by ELISA, confirmed by Immunoblot and with neurological manifestations, between January 2015 - April 2022, in Pedro Hispano Hospital, Matosinhos, Portugal. Results Twenty-one pts: encephalitis (n=5), optic neuropathy (n=5), peripheral neuropathy (n=5), VII cranial nerve palsy (n=4), meningitis (n=3), myelitis (n=2), III cranial nerve palsy (n=1), meningoradiculitis (n=1), radiculitis (n=1) and stroke (n=1); majority female (71.4%) with mean age 50.8 years old. No cases of erythema migrans or arthralgias were observed. ECG performed in 13 pts and only one intraventricular conduction block detected. Lumbar puncture performed in 15 pts, only two presented inflammatory cerebrospinal fluid. Therapeutic strategies: 28 (n=4), 21 (n=5), 14 (n=3) and 10 (n=1) days of ceftriaxone and 28 (n=4), 21 (n=2) and 14 (n=2) days of doxycycline were applied. Complete clinical resolution observed in 6 pts (28.6%) and clinical improvement in 10 (47.6%). One (4.8%) developed post-treatment Lyme disease syndrome and one death occurred, not related with LD. Three pts had no clinical improvement but there was not disease progression, two treated with doxycycline (28 days) and the other one with ceftriaxone (21 days). Conclusion These data show the non-negligible prevalence of NB and the variability of clinical presentations that requires a high suspicion in the differential diagnosis. The heterogeneity of therapeutic regimens reflects the absence of a well-defined treatment and opens this discussion. Disclosures All Authors: No reported disclosures.

Convolutional Neural Network-Based Skin Lesion Classification With Variable Nonlinear Activation Functions

One of the worst forms of skin cancer is melanoma which can be curable if it is diagnosed at an early stage. The earlier the cancer is diagnosed, the better is the outcome. The risk of death from melanoma is directly related to the delay in identifying a lesion. A Deep Learning-based computer diagnosing system can be an automated solution in clinical assessments to overcome this problem. Convolutional Neural Network (CNN) can help to improve the classification rate of skin lesions from dermoscopic images without the need for any human assistance. The linear and nonlinear activation functions act as a node placed at hidden layers or output layers of a Neural Network to play a role in deciding whether that node should pass information to the following layer or not. This critical mathematical mechanism influences the accuracy rate of the CNN. To obtain acceptable performance, CNN requires a large amount of training data. This research shows how fast and effective different types of nonlinear activation functions work on a CNN with limited image datasets. Experimental analysis reveals that the proposed CNN model with parameterized Leaky ReLU function outperforms (97.50% accuracy, 98.00% precision, and 98.00% sensitivity) the same network with distinct nonlinear activation functions for the problem of melanoma recognition by classifying skin lesion into three classes. All experimental studies are carried out using images from PH2 (a dermoscopic image database obtained at the Hospital Pedro Hispano Dermatology Service in Matosinhos, Portugal) and International Skin Imaging Collaboration (ISIC) archive datasets.

Skin Lesion Segmentation in Dermoscopy Imagery

The main purpose of this study is to find an optimum method for segmentation of skin lesion images. In the present world, Skin cancer has proved to be the most deadly disease. The present research paper has developed a model which encompasses two gradations, the first being pre-processing for the reduction of unwanted artefacts like hair, illumination or many other by enhanced technique using threshold and morphological operations to attain higher accuracy and the second being segmentation by using k-mean with optimized Firefly Algorithm (FFA) technique. The online image database from the International Skin Imaging Collaboration (ISIC) archive dataset and dermatology service of Hospital Pedro Hispano (PH2) dataset has been used for input sample images. The parameters on which the proposed method is measured are sensitivity, specificity, dice coefficient, jacquard index, execution time, accuracy, error rate. From the results, authors have observed proposed model gives the average accuracy value of huge number of cancer images using ISIC dataset is 98.9% and using PH2 dataset is 99.1% with minimize average less error rate. It also estimates the dice coefficient value 0.993 using ISIC and 0.998 using PH2 datasets. However, the results for the rest of the parameters remain quite the same. Therefore the outcome of this model is highly reassuring.

1628P Impact of COVID-19 pandemia on the diagnosis of breast cancer in one region of north of Portugal: One year experience

Background: The onset of COVID-19 pandemic forced lockdown and halted breast cancer screening programs. We aimed to investigate the impact of COVID-19 on the new diagnosis and staging of breast cancer. Methods: In this cohort study, we included all patient with new diagnosis of breast cancer who were admitted to our Hospital (Hospital Pedro Hispano, Matosinhos, Portugal), between March 2019 and March 2021. We collected data on baseline clinical conditions such as age, stage at diagnosis and treatment. We created two different groups were created: 1st group- before COVID-19 pandemia (March 1, 2019 to March 16, 2020);2nd group - COVID-19 pandemia (March 17, 2020 to March 31, 2021). A comparative assessment between groups was carried out. Results: Were included 483 patients;n=289 in the 1st group and n= 194 in the 2nd group. The median age was 60 years old in the 1st group and 59 years old in the 2nd group. In the 1st group, 13% patients were diagnosis with ductal in situ carcinoma (DCIS), 51% in stage I, 24% in stage II, 9.5% in stage III and 3% in stage IV. In 2nd group, 9% had DCIS, 30% were in stage I, 40% in stage II, 11% in stage III and 10% in stage IV. Stage at diagnosis was significantly higher in the 2nd group (p< 0.001) This situation was mainly due to tumour size (T). In the 1st group, most patients (n=91;38%) had tumour size between 10 e 20mm (T1c in TNM classification). One the other hand, 40% (n=78) of patients included in the 2nd group had tumour size between 20 e 50mm (T2), with significant differences between them (p=0.004). No difference was found between groups in nodular involvement (p=0.189), with the majority of patients (∼50% in both groups) presenting without nodular involvement (N0 in TMN classification). 10% of patients in 2nd group and 3% in 1st group had metastatic disease at diagnosis, with differences between them (p=0.006). 49% (n=119) of patients in 1st group and 52% (n=100) in the 2nd group were treated with chemotherapy, without differences between those groups. Conclusions: Our results show that during one year after COVID-19 pandemia the incidence of breast cancer decreased, and patients were diagnosis in more advanced stages. This situation could have been related to patient referral to non COVID-19 Hospitals or correspond to a true sub-diagnosis. Legal entity responsible for the study: M. Vilaca. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.

Segmentation of Melanocytic Lesion Images Using Gamma Correction with Clustering of Keypoint Descriptors.

The early detection of skin cancer, especially through the examination of lesions with malignant characteristics, has been reported to significantly decrease the potential fatalities. Segmentation of the regions that contain the actual lesions is one of the most widely used steps for achieving an automated diagnostic process of skin lesions. However, accurate segmentation of skin lesions has proven to be a challenging task in medical imaging because of the intrinsic factors such as the existence of undesirable artifacts and the complexity surrounding the seamless acquisition of lesion images. In this paper, we have introduced a novel algorithm based on gamma correction with clustering of keypoint descriptors for accurate segmentation of lesion areas in dermoscopy images. The algorithm was tested on dermoscopy images acquired from the publicly available dataset of Pedro Hispano hospital to achieve compelling equidistant sensitivity, specificity, and accuracy scores of 87.29%, 99.54%, and 96.02%, respectively. Moreover, the validation of the algorithm on a subset of heavily noised skin lesion images collected from the public dataset of International Skin Imaging Collaboration has yielded the equidistant sensitivity, specificity, and accuracy scores of 80.59%, 100.00%, and 94.98%, respectively. The performance results are propitious when compared to those obtained with existing modern algorithms using the same standard benchmark datasets and performance evaluation indices.

The Impact of 24-Hour Ambulatory EEG in the Clinical Approach to Patients with Suspected Epilepsy

Background: The Electroencephalogram (EEG) is the gold standard technique to assess the epileptogenic cortical activity. However, due to the neurophysiology of the bioelectrical signal and the short duration of the routine EEG (rEEG) and sleepdeprived EEG (sEEG) registers, the sensitivity is low. The 24-hour Ambulatory EEG (aEEG), being a prolonged recording, could significantly improve the diagnostic capability, in an attractive cost-effective way. In this study, we analyzed the aEEG of patients with suspected epilepsy, addressing its specificity and sensitivity for the clinical diagnosis of epilepsy. Additionally, we endeavour to ascertain any other possible predictive factors of diagnosis. Methods and Findings: Retrospective observational study enrolling consecutive patients with suspected epilepsy who underwent aEEG between May 2011 and May 2018 at the Neurophysiology Laboratory from Local Health Unit of Matosinhos – Pedro Hispano Hospital. A sample of 83 individuals was obtained, with a mean age of 44.5 years (79 adults and 4 paediatric). aEEG showed a good diagnostic capacity for the clinical diagnosis of epilepsy with a specificity of 97% and sensitivity of 68%. The rate of false-negatives and false-positives was 7% and 5%, respectively. It’s expected that patients with an indication of syncope or loss of consciousness will not have epilepsy diagnosis. Conclusion: The aEEG can be a useful tool to assess patients with suspected epilepsy and unremarkable routine and sleep-deprived EEGs, or in cases of suspected non-epileptic seizures, particularly to exclude the epilepsy diagnosis given its high specificity. This approach can lessen the time required to identify the diagnosis.

A novel and efficient approach for the classification of skin melanoma

Melanoma is a very common type of skin cancer that is an out-of-control growth of the cell. The detection of melanoma is very crucial in clinical practice due to irregularity in the shape of border and difference of tissue. Even though there are several methods proposed in literature, the identification of melanoma is still a complex process. This work proposes a novel Color Local Directional Pattern-based feature extraction technique for the classification of the skin cancer images. Along with the Color Local Directional Pattern this work also uses various descriptors for shape, color and Pyramid Histogram of Oriented Gradients. This method also uses a stacked Restricted Boltzmann Machine in Deep Belief Network, Support Vector Machine and Random Forest classifier for the classification. The experiments used the images from International Skin Imaging Collaboration (ISIC) 2016, ISIC 2017, Pedro Hispano Hospital (PH2) dataset, Dermnet and DermIS datasets. The performance metrics like specificity, sensitivity, F-Score, positive predictive value, accuracy, harmonic mean, and Area Under the Curve (AUC) are used for the experimental analysis. The results show that the proposed work achieves best results while using the combination of texture, color and shape descriptors and stacked RBM in DBN.

Skin lesion segmentation using k-mean and optimized fire fly algorithm

Digital image processing is turning out to be increasingly more significant in the health care field used to diagnose skin cancer. The death rate is increasing by 1% every year due to skin cancer. One of the major causes of casualties due to this cancer is the non-predictability at the early stages. This paper will help in future research work when it comes to early detection of a tumor. In this work, the proposed model comprises of two important steps which are preprocessing and segmentation. In a pre-processing case, unwanted artifacts like hair, illumination, or many other artifacts are reduced by an enhanced technique using threshold and morphological operations and In the second step, segmentation of skin lesion using k-mean segmentation algorithm with optimized firefly algorithm (FFA) technique is used to achieve high accuracy. Input sample images are taken from the International skin imaging collaboration (ISIC) archive dataset and dermatology service of Hospital Pedro Hispano (PH2) dataset which are available online. The results of the proposed method are measured in terms of different parameters. It provides an accuracy of 99.1% and 98.9% using ISIC and PH2 datasets and shows better performance than existing techniques such as K-Mean and K-Mean with Particle Swarm Optimization (PSO). The performance of this research work is, in fact, quite promising.

Skin Cancer Detection Using Support Vector Machine Learning Classification based on Particle Swarm Optimization Capabilities

Skin cancer continues to be a common malignancy that has steadily increased each year. The need for early detection of such skin lesions is critical to preventing further medical complications. The main method for detection of skin cancer is by microscopic examination of skin lesions. Great efforts have been placed to use computer aided technologies for the analysis of skin lesions. In this study, we present a method for an algorithm design using Support Vector Machine (SVM) learning classification based on Particle swarm optimization (PSO) principles in order to improve the accuracy of skin lesion image analysis and classification for further diagnosis. Hospital Pedro Hispano (PH²) dataset with 200 images is used for this study. The method presented here incorporates 46 texture features in order to complete comprehensive image analytics and classification. The proposed method demonstrates an opportunity to explore best possible criteria in image analytics for clinical decision support.

Deep Learning-Based System for Automatic Melanoma Detection

Melanoma is the deadliest form of skin cancer. Distinguishing melanoma lesions from non-melanoma lesions has however been a challenging task. Many Computer Aided Diagnosis and Detection Systems have been developed in the past for this task. They have been limited in performance due to the complex visual characteristics of the skin lesion images which consists of inhomogeneous features and fuzzy boundaries. In this paper, we propose a deep learning-based method that overcomes these limitations for automatic melanoma lesion detection and segmentation. An enhanced encoder-decoder network with encoder and decoder sub-networks connected through a series of skip pathways which brings the semantic level of the encoder feature maps closer to that of the decoder feature maps is proposed for efficient learning and feature extraction. The system employs multi-stage and multi-scale approach and utilizes softmax classifier for pixel-wise classification of melanoma lesions. We devise a new method called Lesion-classifier that performs the classification of skin lesions into melanoma and non-melanoma based on results derived from pixel-wise classification. Our experiments on two well-established public benchmark skin lesion datasets, International Symposium on Biomedical Imaging(ISBI)2017 and Hospital Pedro Hispano (PH2), demonstrate that our method is more effective than some state-of-the-art methods. We achieved accuracy and dice coefficient of 95% and 92% on ISIC 2017 dataset and accuracy and dice coefficient of 95% and 93% on PH2 datasets.

Robust Boundary Segmentation in Medical Images Using a Consecutive Deep Encoder-Decoder Network

Image segmentation is typically used to locate objects and boundaries. It is essential in many clinical applications, such as the pathological diagnosis of hepatic diseases, surgical planning, and postoperative assessment. The segmentation task is hampered by fuzzy boundaries, complex backgrounds, and appearances of objects of interest, which vary considerably. The success of the procedure is still highly dependent on the operator’s skills and the level of hand–eye coordination. Thus, this paper was strongly motivated by the necessity to obtain an early and accurate diagnosis of a detected object in medical images. In this paper, we propose a new polyp segmentation method based on the architecture of a multiple deep encoder–decoder networks combination called CDED-net. The architecture can not only hold multi-level contextual information by extracting discriminative features at different effective fields-of-view and multiple image scales but also learn rich information features from missing pixels in the training phase. Moreover, the network is also able to capture object boundaries by using multiscale effective decoders. We also propose a novel strategy for improving the method’s segmentation performance based on a combination of a boundary-emphasization data augmentation method and a new effective dice loss function. The goal of this strategy is to make our deep learning network available with poorly defined object boundaries, which are caused by the non-specular transition zone between the background and foreground regions. To provide a general view of the proposed method, our network was trained and evaluated on three well-known polyp datasets, CVC-ColonDB, CVC-ClinicDB, and ETIS-Larib PolypDB. Furthermore, we also used the Pedro Hispano Hospital (PH2), ISBI 2016 skin lesion segmentation dataset, and CT healthy abdominal organ segmentation dataset to depict our network’s ability. Our results reveal that the CDED-net significantly surpasses the state-of-the-art methods.

Automatic skin lesions segmentation based on a new morphological approach via geodesic active contour

Automatic disease classification has been one of the most intensively searched in recent years due to the possibility of quickly providing a diagnosis to the patient. In this process, the segmentation of regions of interest of these diseases has a fundamental role in their subsequent classification. With skin lesions segmentation it is no different and in recent years many studies have achieved interesting results, becoming an important tool in aiding the medical diagnosis of skin diseases. In this work, a morphological geodesic active contour segmentation (MGAC) method is proposed with automatic initialization, using mathematical morphology which is a great partial differential equation approximation, with lower computational cost, no stability problems and fully automatic. The proposed method was tested in a stable and well-known dermoscopic images database provided by Pedro Hispano Hospital (PH2) and was compared with both methods that make use of machine learning or deep learning techniques such as fully convolutional networks (FCN), full resolution convolutional networks (FrCN), deep class-specific learning with probability based step-wise integration (DCL-PSI), and others, and also traditional methods like JSEG, statistical region merging (SRM), Level Set, ASLM and others. The MGAC showed good results in all similarity metrics compared in this work like Jaccard Index (86.16%), Dice coefficient (92.09%) and Matthew correlation coefficient (87.52%), and also achieves good results in sensitivity (91.72%), specificity (97.99%), accuracy (94.59%) and F-measure (93.82%). Thus, the proposed method presented better results in relation to all these metrics when compared to the traditional methods and still presented better results in relation to the methods that use machine learning or deep learning techniques in Jaccard Index, Dice coefficient and specificity. This confirm that the MGAC can efficiently segment skin lesions, presenting great potential to be applied in the aid of the medical diagnosis.

Automatic histologically-closer classification of skin lesions

According to the American Cancer Society, melanoma is one of the most common types of cancer in the world. In 2017, approximately 87,110 new cases of skin cancer were diagnosed in the United States alone. A dermatoscope is a tool that captures lesion images with high resolution and is one of the main clinical tools to diagnose, evaluate and monitor this disease. This paper presents a new approach to classify melanoma automatically using structural co-occurrence matrix (SCM) of main frequencies extracted from dermoscopy images. The main advantage of this approach consists in transform the SCM in an adaptive feature extractor improving his power of discrimination using only the image as parameter. The images were collected from the International Skin Imaging Collaboration (ISIC) 2016, 2017 and Pedro Hispano Hospital (PH2) datasets. Specificity (Spe), sensitivity (Sen), positive predictive value, F Score, Harmonic Mean, accuracy (Acc) and area under the curve (AUC) were used to verify the efficiency of the SCM. The results show that the SCM in the frequency domain work automatically, where it obtained better results in comparison with local binary patterns, gray-level co-occurrence matrix and invariant moments of Hu as well as compared with recent works with the same datasets. The results of the proposed approach were: Spe 95.23%, 92.15% and 99.4%, Sen 94.57%, 89.9% and 99.2%, Acc 94.5%, 89.93% and 99%, and AUC 92%, 90% and 99% in ISIC 2016, 2017 and PH2 datasets, respectively.

Evaluation Methodology between Globalization and Localization Features Approaches for Skin Cancer Lesions Classification

Huge efforts have been put in the developing of diagnostic methods to skin cancer disease. In this paper, two different approaches have been addressed for detection the skin cancer in dermoscopy images. The first approach uses a global method that uses global features for classifying skin lesions, whereas the second approach uses a local method that uses local features for classifying skin lesions. The aim of this paper is selecting the best approach for skin lesion classification. The dataset has been used in this paper consist of 200 dermoscopy images from Pedro Hispano Hospital (PH2). The achieved results are; sensitivity about 96%, specificity about 100%, precision about 100%, and accuracy about 97% for globalization approach while, sensitivity about 100%, specificity about 100%, precision about 100%, and accuracy about 100% for Localization Approach, these results showed that the localization approach achieved acceptable accuracy and better than globalization approach for skin cancer lesions classification.