The pulmonologist's point of view on lung infiltrates in haematological malignancies

Feature Extraction Based on Local Histogram with Unequal Bins and a Recurrent Neural Network for the Diagnosis of Kidney Diseases from CT Images

Kidney disease remains one of the most common ailments worldwide, with cancer being one of its most common forms. Early diagnosis can significantly increase the good prognosis for the patient. The development of an artificial intelligence-based system to assist in kidney cancer diagnosis is crucial because kidney illness is a global health concern, and there are limited nephrologists qualified to evaluate kidney cancer. Diagnosing and categorising different forms of renal failure presents the biggest treatment hurdle for kidney cancer. Thus, this article presents a novel method for detecting and classifying kidney cancer subgroups in Computed Tomography (CT) images based on an asymmetric local statistical pixel distribution. In the first step, the input image is non-overlapping windowed, and a statistical distribution of its pixels in each cancer type is built. Then, the method builds the asymmetric statistical distribution of the image's gradient pixels. Finally, the cancer type is identified by applying the two built statistical distributions to a Deep Neural Network (DNN). The proposed method was evaluated using a dataset collected and authorised by the Dhaka Central International Medical Hospital in Bangladesh, which includes 12,446 CT images of the whole abdomen and urogram, acquired with and without contrast. Based on the results, it is possible to confirm that the proposed method outperformed state-of-the-art methods in terms of the usual correctness criteria. The accuracy of the proposed method for all kidney cancer subtypes presented in the dataset was 99.89%, which is promising.

Kidney segmentation from computed tomography images using deep neural network

BACKGROUND

The precise segmentation of kidneys and kidney tumors can help medical specialists to diagnose diseases and improve treatment planning, which is highly required in clinical practice. Manual segmentation of the kidneys is extremely time-consuming and prone to variability between different specialists due to their heterogeneity. Because of this hard work, computational techniques, such as deep convolutional neural networks, have become popular in kidney segmentation tasks to assist in the early diagnosis of kidney tumors. In this study, we propose an automatic method to delimit the kidneys in computed tomography (CT) images using image processing techniques and deep convolutional neural networks (CNNs) to minimize false positives.

METHODS

The proposed method has four main steps: (1) acquisition of the KiTS19 dataset, (2) scope reduction using AlexNet, (3) initial segmentation using U-Net 2D, and (4) false positive reduction using image processing to maintain the largest elements (kidneys).

RESULTS

The proposed method was evaluated in 210 CTs from the KiTS19 database and obtained the best result with an average Dice coefficient of 96.33%, an average Jaccard index of 93.02%, an average sensitivity of 97.42%, an average specificity of 99.94% and an average accuracy of 99.92%. In the KiTS19 challenge, it presented an average Dice coefficient of 93.03%.

CONCLUSION

In our method, we demonstrated that the kidney segmentation problem in CT can be solved efficiently using deep neural networks to define the scope of the problem and segment the kidneys with high precision and with the use of image processing techniques to reduce false positives.

[Post-transplant pulmonary complications: Guidelines from the francophone Society of bone marrow transplantation and cellular therapy (SFGM-TC)]

Pulmonary complications after allogeneic hematopoietic stem cell transplantation occur frequently (30-75%), vary in severity, and sometimes prove lethal. They may occur at an early stage post-transplant before D100 but may also surface later. Etiological support for these complications has shown a beneficial impact on survival. When faced with early complications, non-invasive tests, scans, and microbiological tests must be rapidly implemented. In the majority of cases, these tests facilitate diagnosis. In cases where microbiological non-invasive tests are negative, and the patient shows a steady respiratory condition, bronchoalveolar lavage can be effective if it is implemented in the first four days following the onset of pulmonary symptoms. This diagnostic approach should in no way occlude the introduction of broad-spectrum antibiotics in these profoundly immunocompromised patients. Later pulmonary complications are the most often not infectious. They include different anatomo-clinical conditions: cryptogenic organizing pneumonia; interstitial lung disease; idiopathic pleuroparenchymal fibroelastosis. Vascular disorders may include hypertension, thrombotic microangiopathy, venous thromboembolism, and pleural effusions. These conditions must be monitored using RFE (respiratory functional exploration) which allows early detection and therapeutic intervention. A combination of RFE and thoracic radiology scans will provide diagnostic assessment. Bronchoalveolar lavage is indicated when an infection is suspected or before systemic corticosteroid therapy. A lung biopsy should be discussed on a case-by-case basis, such as in cases of interstitial pulmonary disorders.

Diagnostic Evaluation of Pulmonary Abnormalities in Patients with Hematologic Malignancies and Hematopoietic Cell Transplantation

Pulmonary complications (PC) of hematologic malignancies and their treatments are common causes of morbidity and mortality. Early diagnosis is challenging due to host risk factors, clinical instability, and provider preference. Delayed diagnosis impairs targeted treatment and may contribute to poor outcomes. An integrated understanding of clinical risk and radiographic patterns informs a timely approach to diagnosis and treatment. There is little prospective evidence guiding optimal modality and timing of minimally invasive lung sampling; however, a low threshold for diagnostic bronchoscopy during the first 24 to 72 hours after presentation should be a guiding principle in high-risk patients.

Early Lung Computed Tomography Scan after Allogeneic Hematopoietic Stem Cell Transplantation

A lung computed tomography (CT) scan is essential for diagnosing lung diseases in hematopoietic stem cell transplantation (HSCT) recipients. As a result, lung CT scans are increasingly prescribed in the early phase after allogeneic HSCT, with no assessment of the added value for global patient management. Among 250 patients who underwent allogeneic HSCT in our center over a 2-year period, we evaluated 68 patients who had at least 1 lung CT scan within the first 30 days post-transplantation. The median interval between allogeneic HSCT and lung CT scan was 8.5 days. Patients who underwent an early lung CT scan were more immunocompromised and had a more severe course. Fever was the main indication for the CT scan (78%). The lung CT scan was abnormal in 52 patients, including 17 patients who had an abnormal pre-HSCT CT scan. A therapeutic change was noted in 37 patients (54%) within 24 hours after the lung CT scan. The main changes included the introduction of corticosteroids (n = 23; 62%), especially in patients with a normal CT scan (89%). In univariate models, we found that a normal pretransplantation CT scan (P = .002), the absence of either dyspnea (P = .029) or hypoxemia (P = .015), and a serum C-reactive protein level <10 mg/L (P = .004) were associated with a normal post-HSCT lung CT scan. We found that the association of these variables could predict the normality of early post-HSCT lung CT scans. Pretransplantation lung CT scans are useful for the interpretation of subsequent lung CT scans following allogeneic HSCT, which are frequently abnormal. Early post-HSCT lung CT scans are helpful in patient management, but prescriptions could be more targeted.

Human metapneumovirus pneumonia in patients with hematological malignancies

•

25% of hematological patients with a positive HMPV test have pneumonia.

•

HMPV pneumonia can occur in the course of several hematological conditions.

•

HMPV can cause pneumonia as a single pathogen.

•

Lung HRCT scan may be suggestive of HMPV pneumonia.

•

The outcome of HMPV pneumonia is good despite no antiviral treatment.

Background

Human metapneumovirus (HMPV) has recently emerged as a cause of respiratory infections in hematological patients. Clinical data are lacking to guide the management of HMPV pneumonias.

Objectives

To characterize the clinical and radiographic presentation and outcome of HMPV pneumonias diagnosed in hematological patients.

Study design

We screened the patients with a positive HMPV respiratory test in two French teaching hospitals between 2007 and 2011. Among them, the medical charts from the hematological patients who presented with HMPV pneumonia were reviewed.

Results

Among the 54 patients with several underlying hematological conditions who were positive for HMPV, we found 13 cases of HMPV pneumonias. HMPV could be the cause of pneumonia as a single pathogen without associated upper respiratory infection. Centrilobular nodules were constant on lung computed tomography scans. No patients died despite the absence of administration of antiviral treatments.

Conclusions

Our data provide further insights in the diagnosis and management of HMPV pneumonias in this setting.