Attention-based deep learning framework for automatic fundus image processing to aid in diabetic retinopathy grading.

BACKGROUND AND OBJECTIVE: Early detection and grading of Diabetic Retinopathy (DR) is essential to determine an adequate treatment and prevent severe vision loss. However, the manual analysis of fundus images is time consuming and DR screening programs are challenged by the availability of human graders. Current automatic approaches for DR grading attempt the joint detection of all signs at the same time. However, the classification can be optimized if red lesions and bright lesions are independently processed since the task gets divided and simplified. Furthermore, clinicians would greatly benefit from explainable artificial intelligence (XAI) to support the automatic model predictions, especially when the type of lesion is specified. As a novelty, we propose an end-to-end deep learning framework for automatic DR grading (5 severity degrees) based on separating the attention of the dark structures from the bright structures of the retina. As the main contribution, this approach allowed us to generate independent interpretable attention maps for red lesions, such as microaneurysms and hemorrhages, and bright lesions, such as hard exudates, while using image-level labels only. METHODS: Our approach is based on a novel attention mechanism which focuses separately on the dark and the bright structures of the retina by performing a previous image decomposition. This mechanism can be seen as a XAI approach which generates independent attention maps for red lesions and bright lesions. The framework includes an image quality assessment stage and deep learning-related techniques, such as data augmentation, transfer learning and fine-tuning. We used the architecture Xception as a feature extractor and the focal loss function to deal with data imbalance. RESULTS: The Kaggle DR detection dataset was used for method development and validation. The proposed approach achieved 83.7 % accuracy and a Quadratic Weighted Kappa of 0.78 to classify DR among 5 severity degrees, which outperforms several state-of-the-art approaches. Nevertheless, the main result of this work is the generated attention maps, which reveal the pathological regions on the image distinguishing the red lesions and the bright lesions. These maps provide explainability to the model predictions. CONCLUSIONS: Our results suggest that our framework is effective to automatically grade DR. The separate attention approach has proven useful for optimizing the classification. On top of that, the obtained attention maps facilitate visual interpretation for clinicians. Therefore, the proposed method could be a diagnostic aid for the early detection and grading of DR.

Post-Chemoradiation Metastatic, Persistent and Resistant Nodes in Locally Advanced Rectal Cancer: Metrics and Their Impact on Long-Term Outcome.

BACKGROUND: The purpose of this study was to evaluate the long-term oncological progression pattern of locally advanced rectal cancer patients with post-neoadjuvant nodal metastatic disease (ypN+) and correlate potential prognostic features associated with proven radiochemoresistant nodal biology. METHODS: Individual patient data (100 variables) from a 20-year consecutive single-institution multidisciplinary experience (1995-2015), delivering multimodal therapy to rectal cancer patient candidates for radical treatment, including a neoadjuvant component and surgical resection with or without intraoperative radiotherapy followed by optional adjuvant chemotherapy. The ypN+ disease data was registered in the context of initial staging categories post-neoadjuvant T status (ypT). RESULTS: Data on 487 patients showed histologically confirmed diagnoses of metastatic nodal disease in 108 specimens (ypN+, 22.1). There was a significant age difference (p = 0.009) between the ypN groups: age ≥ 65 was 57.6% in pN0 and 43.5% in ypN+ and patients aged < 65 constituted 42.4% of pN0 and 56.5% of ypN+. According to the clinical stage there were statistically significant differences (p = 0.001) in the categories' distribution: ypN+ patients 10.8% were stage II and 89.2% were stage III. Univariant analysis on outcome variables showed statistically significant differences in overall survival at 7 years (63.8% vs. 55.7%, p = 0.016) disease-free survival (DFS) (78% vs. 53.8%, p = 0.000) and local recurrence-free survival (LRFS) (93.6% vs. 84%, p = 0.002). CONCLUSIONS: The presence of nodal metastases (ypN+) after neoadjuvant therapy containing long-course pelvic irradiation severely impacts the long-term outcome for patients with locally advanced rectal cancer and correlates with multiple clinical and therapeutic variable metrics. Implementation of local and systemic therapies should be adapted and intensified in relation to the finding of ypN+ category in surgical specimens.

Prevalencia de enfermedades reumáticas en población adulta en España (estudio EPISER 2016). Objetivos y metodología / Prevalence of rheumatic diseases in adult population in Spain (EPISER 2016 study): Aims and methodology

Objetivos: Describir la metodología del estudio de prevalencia de las enfermedades reumáticas en la población adulta en España, EPISER 2016, así como sus fortalezas y limitaciones. El objetivo del proyecto es estimar la prevalencia de artritis reumatoide (AR), artropatía psoriásica (APs), espondilitis anquilosante (EA), lupus eritematoso sistémico (LES), síndrome de Sjögren (SS), artrosis (de rodilla, cadera, manos, columna cervical y lumbar), fibromialgia, gota y fractura osteoporótica clínica. Material y método: Estudio transversal multicéntrico de base poblacional en el que participan 45 municipios de las 17 comunidades autónomas. La población de referencia está compuesta por adultos de 20 o más años residentes en España. La recogida de información se llevará a cabo mediante encuesta telefónica empleando el sistema Computer Assisted Telephone Interview (CATI). Las sospechas diagnósticas y los diagnósticos autorreferidos serán estudiadas por reumatólogos del hospital de referencia de los municipios seleccionados. Análisis estadístico: se calcularán las prevalencias de enfermedades reumáticas mediante estimadores y sus IC del 95%. Se calcularán factores de ponderación en función de la probabilidad de selección en cada una de las etapas del muestreo. Se tendrá en cuenta la distribución de la población en España según datos del Instituto Nacional de Estadística. Conclusiones: Los cambios sociodemográficos y en hábitos de vida durante los últimos 16 años justifican la realización de EPISER 2016. El estudio ofrecerá datos actualizados de prevalencia en AR, EA, APs, LES, SS, artrosis, fibromialgia, gota y fractura osteoporótica clínica. Los resultados permitirán comparar los datos con estudios de otros países y con el EPISER 2000

Aims: To describe the methodology of the EPISER 2016 (study of the prevalence of rheumatic diseases in adult population in Spain), as well its strengths and limitations. The aim of this study is to estimate the prevalence of rheumatoid arthritis (RA), psoriatic arthritis (PsA), ankylosing spondylitis (AS), systemic lupus erythematosus (SLE), Sjögren's syndrome (SS), osteoarthritis (knee, hip, hands, and cervical and lumbar spine), fibromyalgia, gout and clinical osteoporotic fracture. Material and method: Population-based, multicenter, cross-sectional study, with the participation of 45 municipalities in the 17 Spanish autonomous communities. The reference population will consist of adults aged 20 years and over residing in Spain. A computer-assisted telephone interview (CATI) system will be used for data collection. Diagnostic suspicions and diagnoses received by the participants will be studied by rheumatologists in the referral hospitals in the selected municipalities. Statistical analysis: the prevalence of the rheumatic diseases will be calculated using estimators and their 95% confidence intervals. Weights will be calculated in each of the sampling stages in accordance with the probability of selection. The distribution of the population in Spain will be obtained from the Spanish Statistics Institute. Conclusions: Sociodemographic and lifestyle changes over the last 16 years justify EPISER 2016. This study will provide current data about the prevalences of RA, AS, PsA, SLE, SS, osteoarthritis, fibromyalgia, gout and clinical osteoporotic fracture. The results will allow comparisons with studies from other countries and EPISER 2000

Prevalence of rheumatic diseases in adult population in Spain (EPISER 2016 study): Aims and methodology. / Prevalencia de enfermedades reumáticas en población adulta en España (estudio EPISER 2016). Objetivos y metodología.

AIMS: To describe the methodology of the EPISER 2016 (study of the prevalence of rheumatic diseases in adult population in Spain), as well its strengths and limitations. The aim of this study is to estimate the prevalence of rheumatoid arthritis (RA), psoriatic arthritis (PsA), ankylosing spondylitis (AS), systemic lupus erythematosus (SLE), Sjögren's syndrome (SS), osteoarthritis (knee, hip, hands, and cervical and lumbar spine), fibromyalgia, gout and clinical osteoporotic fracture. MATERIAL AND METHOD: Population-based, multicenter, cross-sectional study, with the participation of 45 municipalities in the 17 Spanish autonomous communities. The reference population will consist of adults aged 20 years and over residing in Spain. A computer-assisted telephone interview (CATI) system will be used for data collection. Diagnostic suspicions and diagnoses received by the participants will be studied by rheumatologists in the referral hospitals in the selected municipalities. STATISTICAL ANALYSIS: the prevalence of the rheumatic diseases will be calculated using estimators and their 95% confidence intervals. Weights will be calculated in each of the sampling stages in accordance with the probability of selection. The distribution of the population in Spain will be obtained from the Spanish Statistics Institute. CONCLUSIONS: Sociodemographic and lifestyle changes over the last 16 years justify EPISER 2016. This study will provide current data about the prevalences of RA, AS, PsA, SLE, SS, osteoarthritis, fibromyalgia, gout and clinical osteoporotic fracture. The results will allow comparisons with studies from other countries and EPISER 2000.

Mielitis transversa en síndrome de Sjögren y lupus eritematoso sistémico: presentación de 3 casos / Myelitis transverse in Sjögren´s syndrome and systemic Lupus erythematosus: presentation of 3 cases

La mielitis transversa es una inflamación focal poco frecuente de la médula espinal. Su etiología es múltiple y entre ellas se encuentran las enfermedades autoinmunes, incluyendo principalmente el lupus eritematoso sistémico y el síndrome de Sjögren. Su presentación clínica puede ser de forma aguda o subaguda, con peor pronóstico en la mielitis transversa aguda. Un diagnóstico precoz y tratamiento intensivo desde el inicio es de gran importancia en la evolución de este tipo de pacientes. Presentamos 3 casos con mielitis transversa asociados a enfermedades autoinmunes y discutimos sus distintas manifestaciones clínicas, la asociación a autoanticuerpos, las imágenes radiológicas, el tratamiento y el pronóstico (AU)

Transverse myelitis is a rare focal inflammation of the spinal cord. Multiple etiologies have been identified including autoimmune diseases, mainly systemic lupus erythematosus and Sjögren’ syndrome. It can occur in an acute or subacute clinical onset, with the acute presentation having a worse prognosis. An early diagnosis and intensive treatment are important features recommended in these patients. We present three cases with transverse myelitis associated with autoimmune diseases. We discuss different clinical manifestations, association with autoantobodies, radiologic findings, and therapeutic and prognostic issues (AU)

Myelitis transverse in Sjögren's syndrome and systemic lupus erythematosus: presentation of 3 cases.

Transverse myelitis is a rare focal inflammation of the spinal cord. Multiple etiologies have been identified including autoimmune diseases, mainly systemic lupus erythematosus and Sjögren' syndrome. It can occur in an acute or subacute clinical onset, with the acute presentation having a worse prognosis. An early diagnosis and intensive treatment are important features recommended in these patients. We present three cases with transverse myelitis associated with autoimmune diseases. We discuss different clinical manifestations, association with autoantobodies, radiologic findings, and therapeutic and prognostic issues.