Lymphoma triage from H&E using AI for improved clinical management

Article Computer Science, Artificial Intelligence

Gigapixel end-to-end training using streaming and attention

Stephan Dooper et al.

Summary: This paper proposes a method called StreamingCLAM, which trains a convolutional neural network on gigapixel images and achieves good results. The model is able to detect metastatic breast cancer and MYC-gene translocation in B-cell lymphoma, and offers interpretability through the attention mechanism.

MEDICAL IMAGE ANALYSIS (2023)

Add to Collection

Article Oncology

Global patterns of Hodgkin lymphoma incidence and mortality in 2020 and a prediction of the future burden in 2040

Deependra Singh et al.

Summary: This study examines the global patterns of Hodgkin lymphoma in 2020 and predicts the future incidence and mortality burden. The findings show that males have consistently higher incidence and mortality rates across different regions. Incidence rates vary significantly by regions, with the highest rates observed in Southern Europe, while mortality rates are elevated in Western Asia and Northern Africa. The study predicts an increase in the number of cases by 2040.

INTERNATIONAL JOURNAL OF CANCER (2022)

Add to Collection

Article Oncology

Global patterns of non-Hodgkin lymphoma in 2020

Allini Mafra et al.

Summary: Global patterns of non-Hodgkin lymphoma (NHL) in 2020 were evaluated using the GLOBOCAN 2020 database. The study found that Eastern Asia, Northern America, and South-Central Asia had the highest number of NHL cases. Incidence rates were higher in men than in women, with the highest rates in Australia, New Zealand, Northern America, Northern Europe, and Western Europe. The highest mortality rates were found in regions in Africa, Western Asia, and Oceania.

INTERNATIONAL JOURNAL OF CANCER (2022)

Add to Collection

Article Oncology

Long-Term Cause-Specific Mortality in Hodgkin Lymphoma Patients

Simone de Vries et al.

Summary: Compared with the general population, Hodgkin lymphoma survivors have a substantially reduced life expectancy. Therefore, optimal selection of patients for primary CT is crucial, weighing risks of HL relapse and long-term toxicity.

JNCI-JOURNAL OF THE NATIONAL CANCER INSTITUTE (2021)

Add to Collection

Article Engineering, Biomedical

Data-efficient and weakly supervised computational pathology on whole-slide images

Ming Y. Lu et al.

Summary: The CLAM method utilizes attention-based learning to identify subregions with high diagnostic value for accurate classification of whole-slide images. It can localize well-known morphological features without the need for spatial labels, outperforming standard weakly supervised classification algorithms, and adapt to independent test cohorts, smartphone microscopy, and varying tissue content.

NATURE BIOMEDICAL ENGINEERING (2021)

Add to Collection

Article Oncology

Deep Learning for the Classification of Non-Hodgkin Lymphoma on Histopathological Images

Georg Steinbuss et al.

Summary: Histopathological examination of lymph node specimens can help detect hematological diseases. AI algorithms can be used to automatically detect different lymphoma subtypes with high accuracy, showing the potential for routine diagnostic applications.

CANCERS (2021)

Add to Collection

Article Hematology

Use of Core-Needle Biopsy for the Diagnosis of Malignant Lymphomas in Clinical Practice

Yuta Ito et al.

Summary: The utility of CNB in lymphoma diagnosis has been increasing over the years, with CNB being more frequently opted for biopsies of superficial, abdominal, or anterior mediastinal lesions. The incidence of complications with CNB is comparable to that of EB, while the time to biopsy is shorter with CNB.

ACTA HAEMATOLOGICA (2021)

Add to Collection

Article Pathology