Computationally derived cytological image markers for predicting risk of relapse in acute myeloid leukemia patients following bone marrow transplantation

Arabyarmohammadi, Sara; Zhang, Zelin; Leo, Patrick; Firouznia, Marjan; Janowczyk, Andrew; Li, Haojia; Braman, Nathaniel; Bera, Kaustav; Nezami, Behtash Ghazi; Meyerson, Howard; Xu, Jun; Metheny, Leland; Madabhushi, Anant

doi:10.1117/12.2549730

Cited by 3 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The training patches were subsequently generated from each DWMSI with non-overlapping sliding windows of 512X512 pixel sliding across each ROI ( 19 ). The training patches were then fed into deep semantic segmentation model based on conditional generative adversarial networks (cGAN) ( 20 , 21 ) for training. The parameters were fixed after the training procedure and then were used in the validation cohort of DWMSIs ( n = 41).…”

Section: Methodsmentioning

confidence: 99%

Pros and Cons: High Proportion of Stromal Component Indicates Better Prognosis in Patients With Pancreatic Ductal Adenocarcinoma—A Research Based on the Evaluation of Whole-Mount Histological Slides

Wang

Jiang

et al. 2020

Front. Oncol.

Self Cite

View full text Add to dashboard Cite

The study aimed to investigate the potential of tumor–stroma ratio (TSR) on digitalized whole-mount histopathology to predict prognosis in patients with pancreatic ductal adenocarcinoma (PDAC). The effectiveness were evaluated through internal validation. Data were retrospectively collected from consecutive patients who underwent primary pancreatic resection from December 2016 to August 2017 (developing cohort) and from September 2017 to April 2018 (validation cohort). Digitalized whole-mount slide images were used to evaluate TSR by both pathologists and a computerized model based on Conditional Generative Adversarial Model (cGAN), respectively. TSR>1 and ≤ 1 denoted low and high stromal component. Logistic regression analysis revealed intratumoral necrosis and R1 independently associated with low stromal component in the developing cohort. Cox regression analysis revealed tumor–node–metastasis (TNM) stage [II vs. I: hazard ratio (HR), 2.584; 95% CI, 1.386–4.819; P = 0.003; III vs. I: HR, 4.384; 95% CI, 2.285–8.411; P < 0.001], stromal component (low vs. high: HR, 1.876; 95% CI, 1.227–2.870; P = 0.004), tumor grade (G3 vs. G1/2: HR, 2.124; 95% CI, 1.419–3.179; P < 0.001), and perineural invasion (with vs. without: HR, 2.147; 95% CI, 1.187–3.883; P = 0.011) were independent prognostic factors in the developing cohort. Stromal component categories could classify patients into subgroups within TNM stages I, II, and III based on over survival. All results were validated in the validation cohort. The weighted kappa value for categorical assessments between pathologists' evaluation and computer-aided evaluation was 0.804 (95% CI, 0.573–0.951). TSR represents a simple and reliable metric for combining the prognostic value of TNM stage in patients with PDAC.

show abstract

Section: Methodsmentioning

confidence: 99%

Pros and Cons: High Proportion of Stromal Component Indicates Better Prognosis in Patients With Pancreatic Ductal Adenocarcinoma—A Research Based on the Evaluation of Whole-Mount Histological Slides

Wang

Jiang

et al. 2020

Front. Oncol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, Arab Yarmohammadi et al [21] proposed an algorithm to detect post-HSCT relapse in AML patients using automated image analysis. Bone marrow Wright-Giemsa aspirate slides were collected from 39 AML patients and a deep learning algorithm was employed to segment myeloblasts (i.e., a cell type in bone marrow to characterize AML).…”

Section: Post-hsct Complicationsmentioning

confidence: 99%

A Systematic Review of Machine Learning Techniques in Hematopoietic Stem Cell Transplantation (HSCT)

Gupta

Braun

Chowdhury

et al. 2020

Sensors

View full text Add to dashboard Cite

Machine learning techniques are widely used nowadays in the healthcare domain for the diagnosis, prognosis, and treatment of diseases. These techniques have applications in the field of hematopoietic cell transplantation (HCT), which is a potentially curative therapy for hematological malignancies. Herein, a systematic review of the application of machine learning (ML) techniques in the HCT setting was conducted. We examined the type of data streams included, specific ML techniques used, and type of clinical outcomes measured. A systematic review of English articles using PubMed, Scopus, Web of Science, and IEEE Xplore databases was performed. Search terms included “hematopoietic cell transplantation (HCT),” “autologous HCT,” “allogeneic HCT,” “machine learning,” and “artificial intelligence.” Only full-text studies reported between January 2015 and July 2020 were included. Data were extracted by two authors using predefined data fields. Following PRISMA guidelines, a total of 242 studies were identified, of which 27 studies met the inclusion criteria. These studies were sub-categorized into three broad topics and the type of ML techniques used included ensemble learning (63%), regression (44%), Bayesian learning (30%), and support vector machine (30%). The majority of studies examined models to predict HCT outcomes (e.g., survival, relapse, graft-versus-host disease). Clinical and genetic data were the most commonly used predictors in the modeling process. Overall, this review provided a systematic review of ML techniques applied in the context of HCT. The evidence is not sufficiently robust to determine the optimal ML technique to use in the HCT setting and/or what minimal data variables are required.

show abstract

A systematic review of the predicted outcomes related to hematopoietic stem cell transplantation: focus on applied machine learning methods’ performance

Taheriyan

Nodehi

Kalhori

et al. 2022

Expert Review of Hematology

View full text Add to dashboard Cite

Computationally derived cytological image markers for predicting risk of relapse in acute myeloid leukemia patients following bone marrow transplantation

Cited by 3 publications

References 0 publications

Pros and Cons: High Proportion of Stromal Component Indicates Better Prognosis in Patients With Pancreatic Ductal Adenocarcinoma—A Research Based on the Evaluation of Whole-Mount Histological Slides

Pros and Cons: High Proportion of Stromal Component Indicates Better Prognosis in Patients With Pancreatic Ductal Adenocarcinoma—A Research Based on the Evaluation of Whole-Mount Histological Slides

A Systematic Review of Machine Learning Techniques in Hematopoietic Stem Cell Transplantation (HSCT)

A systematic review of the predicted outcomes related to hematopoietic stem cell transplantation: focus on applied machine learning methods’ performance

Contact Info

Product

Resources

About