Comprehensive analysis of clinical images contributions for melanoma classification using convolutional neural networks.

BACKGROUND: Timely diagnosis plays a critical role in determining melanoma prognosis, prompting the development of deep learning models to aid clinicians. Questions persist regarding the efficacy of clinical images alone or in conjunction with dermoscopy images for model training. This study aims to compare the classification performance for melanoma of three types of CNN models: those trained on clinical images, dermoscopy images, and a combination of paired clinical and dermoscopy images from the same lesion. MATERIALS AND METHODS: We divided 914 image pairs into training, validation, and test sets. Models were built using pre-trained Inception-ResNetV2 convolutional layers for feature extraction, followed by binary classification. Training comprised 20 models per CNN type using sets of random hyperparameters. Best models were chosen based on validation AUC-ROC. RESULTS: Significant AUC-ROC differences were found between clinical versus dermoscopy models (0.661 vs. 0.869, p < 0.001) and clinical versus clinical + dermoscopy models (0.661 vs. 0.822, p = 0.001). Significant sensitivity differences were found between clinical and dermoscopy models (0.513 vs. 0.799, p = 0.01), dermoscopy versus clinical + dermoscopy models (0.799 vs. 1.000, p = 0.02), and clinical versus clinical + dermoscopy models (0.513 vs. 1.000, p < 0.001). Significant specificity differences were found between dermoscopy versus clinical + dermoscopy models (0.800 vs. 0.288, p < 0.001) and clinical versus clinical + dermoscopy models (0.650 vs. 0.288, p < 0.001). CONCLUSION: CNN models trained on dermoscopy images outperformed those relying solely on clinical images under our study conditions. The potential advantages of incorporating paired clinical and dermoscopy images for CNN-based melanoma classification appear less clear based on our findings.

Epidemiología digital como herramienta para la evaluación de enfermedades del oído y la vía aérea en Colombia, durante los años 2012-2014: Un estudio ecológico / Digital epidemiology as a tool for the evaluation of ear and airway diseases in Colombia, during the years 2012-2014: An ecological study

Introducción: Las tecnologías de información y la facilidad de acceso a grandes volúmenes de datos "big data" están revolucionando la epidemiología. Objetivo: Analizar datos del sistema integral de Información "SISPRO" del Ministerio de Salud para obtener los motivos más frecuentes de consulta en otorrinolaringología en consulta externa y urgencias. Diseño: Estudio descriptivo ecológico. Metodología: En SISPRO se usaron los códigos CIE-10 para analizar los datos de atención en salud para patologías de oído (H60-H959) y vía aérea superior (J00-J399). Resultados: Los principales motivos de consulta incluyen: infecciones de la vía aérea superior (rinofaringitis, faringitis, amigdalitis, otitis media aguda e infecciones no especificadas) con 14.935.729 atenciones, 7.877.189 personas atendidas y un promedio de 1,9 consultas por persona; Rinitis (1.760.055 atenciones, 899.349 personas y 1,96 consultas/persona); Vértigo (1.632.012 atenciones, 686.470 personas y 2,38 consultas/persona); Cerumen impactado (554.744 atenciones, 296.324 personas y 18,76 consultas/persona); Hipoacusia (853.047 atenciones, 366.037 personas y 2,33 consultas/persona). La relación entre el número de atenciones en consulta externa vs urgencias es de 281,66 (hipoacusia), 126,23 (Rinitis), 37,2 (Cerumen impactado), 10,73 (vértigo) y 6,9 (infecciones). Conclusión: La epidemiología digital es una herramienta útil en la práctica de ORL. Las enfermedades infecciosas son la principal causa de atención en ORL.

Introduction: Information technologies and the ease of access to large volumes of data "big data" are revolutionizing epidemiology. Objective: Analyze data from the Comprehensive Information System "SISPRO" of the Ministry of Health, to obtain the most frequent reasons for consultation in otorhinolaryngology in outpatient and emergency departments. Design: ecological descriptive study. Methods: In SISPRO, ICD-10 codes were used to analyze health care data for ear (H60-H959) and upper airway (J00-J399) pathologies. Results: The main reasons for consultation included: upper airway infections (rhinopharyngitis, pharyngitis, tonsillitis, acute otitis media and unspecified infections) with 14,935,729 attentions, 7,877,189 people attended and an average of 1.9 consultations per person; Rhinitis (1,760,055 attentions, 899,349 people and 1.96 consultations / person); Vertigo (1,632,012 attentions, 686,470 people and 2,38 consultations / person; Cerumen impacted (554,744 attentions, 296,324 people and 18,76 consultations / person); Hearing loss (853,047 attentions, 366,037 people and 2,33 consultations / person). The ratio between the number of visits in the outpatient clinic vs. emergencies is 281.66 (hearing loss), 126.23 (rhinitis), 37.2 (impacted earwax), 10.73 (vertigo) and 6.9 (infections). Conclusion: Digital epidemiology is an useful tool in ENT. Infectious diseases are the main cause of consultation in ENT.

Myeloperoxidase is increased in human cerebral aneurysms and increases formation and rupture of cerebral aneurysms in mice.

BACKGROUND AND PURPOSE: Cerebral aneurysm (CA) affects 3% of the population and is associated with hemodynamic stress and inflammation. Myeloperoxidase, a major oxidative enzyme associated with inflammation, is increased in patients with CA, but whether myeloperoxidase contributes to CA is not known. We tested the hypotheses that myeloperoxidase is increased within human CA and is critical for formation and rupture of CA in mice. METHODS: Blood was drawn from the lumen of CAs and femoral arteries of 25 patients who underwent endovascular coiling of CA, and plasma myeloperoxidase concentrations were measured with ELISA. Effects of endogenous myeloperoxidase on CA formation and rupture were studied in myeloperoxidase knockout mice and wild-type (WT) mice using an angiotensin II-elastase induction model of CA. In addition, effects of myeloperoxidase on inflammatory gene expression in endothelial cells were analyzed. RESULTS: Plasma concentrations of myeloperoxidase were 2.7-fold higher within CA than in femoral arterial blood in patients with CA. myeloperoxidase-positive cells were increased in aneurysm tissue compared with superficial temporal artery of patients with CA. Incidence of aneurysms and subarachnoid hemorrhage was significantly lower in myeloperoxidase knockout than in WT mice. In cerebral arteries, proinflammatory molecules, including tumor necrosis factor-α, cyclooxygenase-2 (COX2), chemokine (C-X-C motif) ligand 1 (CXCL1), chemokine (C motif) ligand (XCL1), matrix metalloproteinase (MMP) 8, cluster of differentiation 68 (CD68), and matrix metalloproteinase 13, and leukocytes were increased, and α-smooth muscle actin was decreased, in WT but not in myeloperoxidase knockout mice after induction of CA. Myeloperoxidase per se increased expression of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 in endothelial cells. CONCLUSIONS: These findings suggest that myeloperoxidase may contribute importantly to formation and rupture of CA.

Novel role for endogenous hepatocyte growth factor in the pathogenesis of intracranial aneurysms.

Inflammation plays a key role in formation and rupture of intracranial aneurysms. Because hepatocyte growth factor (HGF) protects against vascular inflammation, we sought to assess the role of endogenous HGF in the pathogenesis of intracranial aneurysms. Circulating HGF concentrations in blood samples drawn from the lumen of human intracranial aneurysms or femoral arteries were compared in 16 patients. Tissue from superficial temporal arteries and ruptured or unruptured intracranial aneurysms collected from patients undergoing clipping (n=10) were immunostained with antibodies to HGF and its receptor c-Met. Intracranial aneurysms were induced in mice treated with PF-04217903 (a c-Met antagonist) or vehicle. Expression of inflammatory molecules was also measured in cultured human endothelial, smooth muscle cells and monocytes treated with lipopolysaccharides in presence or absence of HGF and PF-04217903. We found that HGF concentrations were significantly higher in blood collected from human intracranial aneurysms (1076±656 pg/mL) than in femoral arteries (196±436 pg/mL; P<0.001). HGF and c-Met were detected by immunostaining in superficial temporal arteries and in both ruptured and unruptured human intracranial aneurysms. A c-Met antagonist did not alter the formation of intracranial aneurysms (P>0.05), but significantly increased the prevalence of subarachnoid hemorrhage and decreased survival in mice (P<0.05). HGF attenuated expression of vascular cell adhesion molecule-1 (P<0.05) and E-Selectin (P<0.05) in human aortic endothelial cells. In conclusion, plasma HGF concentrations are elevated in intracranial aneurysms. HGF and c-Met are expressed in superficial temporal arteries and in intracranial aneurysms. HGF signaling through c-Met may decrease inflammation in endothelial cells and protect against intracranial aneurysm rupture.

Angiotensin 1-7 reduces mortality and rupture of intracranial aneurysms in mice.

Angiotensin II (Ang II) stimulates vascular inflammation, oxidative stress, and formation and rupture of intracranial aneurysms in mice. Because Ang 1-7 acts on Mas receptors and generally counteracts deleterious effects of Ang II, we tested the hypothesis that Ang 1-7 attenuates formation and rupture of intracranial aneurysms. Intracranial aneurysms were induced in wild-type and Mas receptor-deficient mice using a combination of Ang II-induced hypertension and intracranial injection of elastase in the basal cistern. Mice received elastase+Ang II alone or a combination of elastase+Ang II+Ang 1-7. Aneurysm formation, prevalence of subarachnoid hemorrhage, mortality, and expression of molecules involved in vascular injury were assessed. Systolic blood pressure was similar in mice receiving elastase+Ang II (mean±SE, 148±5 mm Hg) or elastase+Ang II+Ang 1-7 (144±5 mm Hg). Aneurysm formation was also similar in mice receiving elastase+Ang II (89%) or elastase+Ang II+Ang 1-7 (84%). However, mice that received elastase+Ang II+Ang 1-7 had reduced mortality (from 64% to 36%; P<0.05) and prevalence of subarachnoid hemorrhage (from 75% to 48%; P<0.05). In cerebral arteries, expression of the inflammatory markers, Nox2 and catalase increased similarly in elastase+Ang II or elastase+Ang II+Ang 1-7 groups. Ang 1-7 increased the expression of cyclooxygenase-2 and decreased the expression of matrix metalloproteinase-9 induced by elastase+Ang II (P<0.05). In Mas receptor-deficient mice, systolic blood pressure, mortality, and prevalence of subarachnoid hemorrhage were similar (P>0.05) in groups treated with elastase+Ang II or elastase+Ang II+Ang 1-7. The expression of Mas receptor was detected by immunohistochemistry in samples of human intracranial arteries and aneurysms. In conclusion, without attenuating Ang II-induced hypertension, Ang 1-7 decreased mortality and rupture of intracranial aneurysms in mice through a Mas receptor-dependent pathway.

Estrogen protects against intracranial aneurysm rupture in ovariectomized mice.

Clinical observations suggest that postmenopausal women have a higher incidence of aneurysmal rupture than premenopausal women. We hypothesize that a relative deficiency in estrogen may increase the risks of aneurysmal growth and subarachnoid hemorrhage in postmenopausal women. We assessed the effects of estrogen and selective estrogen receptor subtype agonists on the development of aneurysmal rupture in ovariectomized female mice. We used an intracranial aneurysm mouse model that recapitulates the key features of human intracranial aneurysms, including spontaneous rupture. Ten- to 12-week-old ovariectomized female mice received treatment with estrogen, nonselective estrogen receptor antagonist, estrogen receptor-α agonist, or estrogen receptor-ß agonist starting 6 days after aneurysm induction so that the treatments affected the development of aneurysmal rupture without affecting aneurysmal formation. Estrogen significantly reduced the incidence of ruptured aneurysms and rupture rates in ovariectomized mice. Nonselective estrogen receptor antagonist abolished the protective effect of estrogen. Although estrogen receptor-α agonist did not affect the incidence of ruptured aneurysms or rupture rates, estrogen receptor-ß agonist prevented aneurysmal rupture without affecting the formation of aneurysms. The protective role of estrogen receptor-ß agonist was abolished by the inhibition of nitric oxide synthase. We showed that estrogen prevented aneurysmal rupture in ovariectomized female mice. The protective effect of estrogen seemed to occur through the activation of estrogen receptor-ß, a predominant subtype of estrogen receptor in human intracranial aneurysms and cerebral arteries.

Impact of ACE2 deficiency and oxidative stress on cerebrovascular function with aging.

BACKGROUND AND PURPOSE: Angiotensin II produces oxidative stress and endothelial dysfunction in cerebral arteries, and angiotensin II type I receptors may play a role in longevity and vascular aging. Angiotensin-converting enzyme type 2 (ACE2) converts angiotensin II to angiotensin (1-7) and thus, may protect against effects of angiotensin II. We hypothesized that ACE2 deficiency increases oxidative stress and endothelial dysfunction in cerebral arteries and examined the role of ACE2 in age-related cerebrovascular dysfunction. METHODS: Endothelial function, expression of angiotensin system components, NADPH oxidase subunits, and proinflammatory cytokines were examined in cerebral arteries from adult (12 months old) and old (24 months old) ACE2 knockout (KO) and wild-type (WT) mice. The superoxide scavenger tempol was used to examine the role of oxidative stress on endothelial function. RESULTS: Vasodilatation to acetylcholine was impaired in adult ACE2 KO (24±6% [mean±SE]) compared with WT mice (52±7%; P<0.05). In old mice, vasodilatation to acetylcholine was impaired in WT mice (29±6%) and severely impaired in ACE2 KO mice (7±5%). Tempol improved endothelial function in adult and old ACE2 KO and WT mice. Aging increased mRNA for tumor necrosis factor-α in WT mice, and significantly increased mRNA levels of NAPDH oxidase 2, p47(phox), and Regulator of calcineurin 1 in both ACE2 KO and WT mice. mRNA levels of angiotensin system components did not change during aging. CONCLUSIONS: ACE2 deficiency impaired endothelial function in cerebral arteries from adult mice and augmented endothelial dysfunction during aging. Oxidative stress plays a critical role in cerebrovascular dysfunction induced by ACE2 deficiency and aging.

Serotonin produces monoamine oxidase-dependent oxidative stress in human heart valves.

Heart valve disease and pulmonary hypertension, in patients with carcinoid tumors and people who used the fenfluramine-phentermine combination for weight control, have been associated with high levels of serotonin in blood. The mechanism by which serotonin induces valvular changes is not well understood. We recently reported that increased oxidative stress is associated with valvular changes in aortic valve stenosis in humans and mice. In this study, we tested the hypothesis that serotonin induces oxidative stress in human heart valves, and examined mechanisms by which serotonin may increase reactive oxygen species. Superoxide (O2*.-) was measured in heart valves from explanted human hearts that were not used for transplantation. (O2*.-) levels (lucigenin-enhanced chemoluminescence) were increased in homogenates of cardiac valves and blood vessels after incubation with serotonin. A nonspecific inhibitor of flavin-oxidases (diphenyliodonium), or inhibitors of monoamine oxidase [MAO (tranylcypromine and clorgyline)], prevented the serotonin-induced increase in (O2*.-). Dopamine, another MAO substrate that is increased in patients with carcinoid syndrome, also increased (O2*.-) levels in heart valves, and this effect was attenuated by clorgyline. Apocynin [an inhibitor of NAD(P)H oxidase] did not prevent increases in (O2*.-) during serotonin treatment. Addition of serotonin to recombinant human MAO-A generated (O2*.-), and this effect was prevented by an MAO inhibitor. In conclusion, we have identified a novel mechanism whereby MAO-A can contribute to increased oxidative stress in human heart valves and pulmonary artery exposed to serotonin and dopamine.