Effective and efficient active learning for deep learning-based tissue image analysis.

MOTIVATION: Deep learning attained excellent results in digital pathology recently. A challenge with its use is that high quality, representative training datasets are required to build robust models. Data annotation in the domain is labor intensive and demands substantial time commitment from expert pathologists. Active learning (AL) is a strategy to minimize annotation. The goal is to select samples from the pool of unlabeled data for annotation that improves model accuracy. However, AL is a very compute demanding approach. The benefits for model learning may vary according to the strategy used, and it may be hard for a domain specialist to fine tune the solution without an integrated interface. RESULTS: We developed a framework that includes a friendly user interface along with run-time optimizations to reduce annotation and execution time in AL in digital pathology. Our solution implements several AL strategies along with our diversity-aware data acquisition (DADA) acquisition function, which enforces data diversity to improve the prediction performance of a model. In this work, we employed a model simplification strategy [Network Auto-Reduction (NAR)] that significantly improves AL execution time when coupled with DADA. NAR produces less compute demanding models, which replace the target models during the AL process to reduce processing demands. An evaluation with a tumor-infiltrating lymphocytes classification application shows that: (i) DADA attains superior performance compared to state-of-the-art AL strategies for different convolutional neural networks (CNNs), (ii) NAR improves the AL execution time by up to 4.3×, and (iii) target models trained with patches/data selected by the NAR reduced versions achieve similar or superior classification quality to using target CNNs for data selection. AVAILABILITY AND IMPLEMENTATION: Source code: https://github.com/alsmeirelles/DADA.

Building Efficient CNN Architectures for Histopathology Images Analysis: A Case-Study in Tumor-Infiltrating Lymphocytes Classification.

Background: Deep learning methods have demonstrated remarkable performance in pathology image analysis, but they are computationally very demanding. The aim of our study is to reduce their computational cost to enable their use with large tissue image datasets. Methods: We propose a method called Network Auto-Reduction (NAR) that simplifies a Convolutional Neural Network (CNN) by reducing the network to minimize the computational cost of doing a prediction. NAR performs a compound scaling in which the width, depth, and resolution dimensions of the network are reduced together to maintain a balance among them in the resulting simplified network. We compare our method with a state-of-the-art solution called ResRep. The evaluation is carried out with popular CNN architectures and a real-world application that identifies distributions of tumor-infiltrating lymphocytes in tissue images. Results: The experimental results show that both ResRep and NAR are able to generate simplified, more efficient versions of ResNet50 V2. The simplified versions by ResRep and NAR require 1.32× and 3.26× fewer floating-point operations (FLOPs), respectively, than the original network without a loss in classification power as measured by the Area under the Curve (AUC) metric. When applied to a deeper and more computationally expensive network, Inception V4, NAR is able to generate a version that requires 4× lower than the original version with the same AUC performance. Conclusions: NAR is able to achieve substantial reductions in the execution cost of two popular CNN architectures, while resulting in small or no loss in model accuracy. Such cost savings can significantly improve the use of deep learning methods in digital pathology. They can enable studies with larger tissue image datasets and facilitate the use of less expensive and more accessible graphics processing units (GPUs), thus reducing the computing costs of a study.

Effective active learning in digital pathology: A case study in tumor infiltrating lymphocytes.

BACKGROUND AND OBJECTIVE: Deep learning methods have demonstrated remarkable performance in pathology image analysis, but they require a large amount of annotated training data from expert pathologists. The aim of this study is to minimize the data annotation need in these analyses. METHODS: Active learning (AL) is an iterative approach to training deep learning models. It was used in our context with a Tumor Infiltrating Lymphocytes (TIL) classification task to minimize annotation. State-of-the-art AL methods were evaluated with the TIL application and we have proposed and evaluated a more efficient and effective AL acquisition method. The proposed method uses data grouping based on imaging features and model prediction uncertainty to select meaningful training samples (image patches). RESULTS: An experimental evaluation with a collection of cancer tissue images shows that: (i) Our approach reduces the number of patches required to attain a given AUC as compared to other approaches, and (ii) our optimization (subpooling) leads to AL execution time improvement of about 2.12×. CONCLUSIONS: This strategy enabled TIL based deep learning analyses using smaller annotation demand. We expect this approach may be used to build other analyses in digital pathology with fewer training samples.

Photoreceptor assessment using adaptive optics in resolved central serous chorioretinopathy.

We present a case of a patient with resolved central serous chorioretinopathy (CSC) in the left eye with persistent mild visual alterations 4 years after the resolution of the disease. Left eye exam revealed a best corrected visual acuity of 20/25 and a slight change of macular pigmentation. Optical coherence tomography revealed only minor irregularities in the topography of retinal pigment epithelium and Bruch's membrane. Adaptive optics (AO) optics demonstrated lower density, spacing, and changes in the photoreceptor mosaic pattern in the left eye than in the right eye, suggesting that CSC may cause damage to cones after clinical recovery. We conclude that AO can provide additional information to assist in the treatment and follow-up of patients with CSC or other macular pathologies.

Photoreceptor assessment using adaptive optics in resolved central serous chorioretinopathy / Avaliação dos fotorreceptores com adaptive optics na coriorretinopatia serosa central cicatrizada

ABSTRACT We present a case of a patient with resolved central serous chorioretinopathy (CSC) in the left eye with persistent mild visual alterations 4 years after the resolution of the disease. Left eye exam revealed a best corrected visual acuity of 20/25 and a slight change of macular pigmentation. Optical coherence tomography revealed only minor irregularities in the topography of retinal pigment epithelium and Bruch's membrane. Adaptive optics (AO) optics demonstrated lower density, spacing, and changes in the photoreceptor mosaic pattern in the left eye than in the right eye, suggesting that CSC may cause damage to cones after clinical recovery. We conclude that AO can provide additional information to assist in the treatment and follow-up of patients with CSC or other macular pathologies.

RESUMO Apresentamos o relato de caso de paciente com coriorretinopatia serosa central (CSC) cicatrizada em olho esquerdo e queixa de discreta alteração visual, mesmo após quatro anos da resolução do quadro. O exame do olho esquerdo apresenta melhor acuidade visual corrigida de 20/25 e discreta alteração de pigmentação macular. Tomografia de coerência óptica (OCT) apresentou apenas pequenas irregularidades em topografia de EPR e Bruch. Foi realizado exame com Adaptive Optics (AO), evidenciando valores inferiores de densidade, espaçamento e alterações no padrão de mosaico dos fotorreceptores em olho esquerdo quando comparado com olho direito, sugerindo que a CSC pode causar danos em cones, mesmo após uma recuperação considerada satisfatória. Concluímos que o AO é uma tecnologia que traz novas informações para auxiliar o tratamento e seguimento dos pacientes com CSC ou outras patologias maculares.

Macular dystrophy associated with Kjellin's syndrome: a case report.

Hereditary spastic paraplegia (HSP) is characterized by weakness and spasticity of the lower extremities. Kjellin's syndrome is a rare syndrome associated with HSP. The syndrome is characterized by the presence of bilateral retinal flecks, similar to the findings in Stargardt disease and fundus flavimaculatus. We report the case of a 34-year-old male who presented with complete features of Kjellin's syndrome, with typical retinal findings observed on multimodal imaging (spectral domain optical coherence tomography [SD-OCT], near-infrared reflectance and autofluorescence imaging). The ophthalmological changes at early stages of the disease may not impair visual acuity. Therefore, the detection of central retinal degeneration requires thorough fundus examination.

Macular dystrophy associated with Kjellin’s syndrome: a case report / Distrofia macular associada à síndrome de Kjellin: um relato de caso

Hereditary spastic paraplegia (HSP) is characterized by weakness and spasticity of the lower extremities. Kjellin’s syndrome is a rare syndrome associated with HSP. The syndrome is characterized by the presence of bilateral retinal flecks, similar to the findings in Stargardt disease and fundus flavimaculatus. We report the case of a 34-year-old male who presented with complete features of Kjellin’s syndrome, with typical retinal findings observed on multimodal imaging (spectral domain optical coherence tomography [SD-OCT], near-infrared reflectance and autofluorescence imaging). The ophthalmological changes at early stages of the disease may not impair visual acuity. Therefore, the detection of central retinal degeneration requires thorough fundus examination.

A paralisia espástica hereditária (HSP) é caracterizada por fraqueza e espasticidade das extremidades inferiores. A síndrome de Kjellin é uma rara associação de HSP com a presença de flecks retinianos similares aos encontrados em pacientes com doença de Stargardt ou fundus flavimaculatus. Descrevemos os achados em imagens multimodais da retina (tomografia de coerência óptica de domínio espectral [SD-OCT], reflectância próxima ao infravermelho e autofluorescência) em um paciente de 34 anos que apresenta conjunto completo de sinais e sintomas da síndrome de Kjellin. As alterações retinianas em estágios iniciais da doença podem aparecer, mesmo sem redução da acuidade visual, e por isso, para detecção da degeneração central da retina, é necessário exame minucioso do fundo de olho.

Diethylstilbestrol 1 mg in the Treatment of Acute Urinary Retention due to Prostatic Obstruction in the Elderly: A Preliminary Study.

Patients who failed a catheter-free trial after acute urinary retention and one week of full dose alpha-blocker and 5-alpha-reductase inhibitor were offered Diethylstilbestrol 1 mg plus Aspirin 100 mg over 4 weeks. Prostate volume, age, serum creatinine, and initial retention drained urine volume were recorded. After excluding cardiovascular morbidity (n = 7), upper urinary tract dilation (n = 3), compromised renal function (n = 2), urinary tract infection (n = 2), neurological diagnosis (n = 2), or preferred immediate channel transurethral resection of prostate (n = 5), 48 of 69 consecutive patients ≥70 years were included. Mean age was 76.6 years (70-84), mean prostate volume 90 cm(3) (42-128), and mean follow-up 204 days; 58% (28/48) were passing urine and 42% (20/48) were catheter dependent after 4 weeks Diethylstilbestrol trial. Mean age and drained urine volume of catheter dependent patients were 82.4 years and 850 mL compared with 74.6 years and 530 mL in catheter-free men, respectively. Age and drained urine volume were independent predictors of catheter-free trial (both P < 0.01). Seventy-five percent (6/8) of patients 80 years and older were catheter dependent. Transient nipple/breast tenderness and gynecomastia were the only adverse effects reported by 21% (10/48) and 4% (2/48), respectively. No patient presented severe complications.