Submitted Successfully!
To reward your contribution, here is a gift for you: A free trial for our video production service.
Thank you for your contribution! You can also upload a video entry or images related to this topic.
Version Summary Created by Modification Content Size Created at Operation
1
Basel Elsayed
 -- 2462 2024-02-20 13:16:37 |
2 Reference format revised.
Lindsay Dong
 + 4 word(s) 2466 2024-02-21 02:12:54 |

Video Upload Options

Do you have a full video?
Send video materials Upload full video

Confirm

Are you sure to Delete?
Yes No
Cite
If you have any further questions, please contact Encyclopedia Editorial Office.
Elshoeibi, A.M.; Badr, A.; Elsayed, B.; Metwally, O.; Elshoeibi, R.; Elhadary, M.R.; Elshoeibi, A.; Attya, M.A.; Khadadah, F.; Alshurafa, A.; et al. Integrating AI and ML in Myelodysplastic Syndrome Diagnosis. Encyclopedia. Available online: https://encyclopedia.pub/entry/55226 (accessed on 18 May 2024).
Elshoeibi AM, Badr A, Elsayed B, Metwally O, Elshoeibi R, Elhadary MR, et al. Integrating AI and ML in Myelodysplastic Syndrome Diagnosis. Encyclopedia. Available at: https://encyclopedia.pub/entry/55226. Accessed May 18, 2024.
Elshoeibi, Amgad Mohamed, Ahmed Badr, Basel Elsayed, Omar Metwally, Raghad Elshoeibi, Mohamed Ragab Elhadary, Ahmed Elshoeibi, Mohamed Amro Attya, Fatima Khadadah, Awni Alshurafa, et al. "Integrating AI and ML in Myelodysplastic Syndrome Diagnosis" Encyclopedia, https://encyclopedia.pub/entry/55226 (accessed May 18, 2024).
Elshoeibi, A.M., Badr, A., Elsayed, B., Metwally, O., Elshoeibi, R., Elhadary, M.R., Elshoeibi, A., Attya, M.A., Khadadah, F., Alshurafa, A., Alhuraiji, A., & Yassin, M. (2024, February 20). Integrating AI and ML in Myelodysplastic Syndrome Diagnosis. In Encyclopedia. https://encyclopedia.pub/entry/55226
Elshoeibi, Amgad Mohamed, et al. "Integrating AI and ML in Myelodysplastic Syndrome Diagnosis." Encyclopedia. Web. 20 February, 2024.
Integrating AI and ML in Myelodysplastic Syndrome Diagnosis
Edit
This entry is adapted from the peer-reviewed paper 10.3390/cancers16010065

Myelodysplastic syndrome (MDS) is composed of diverse hematological malignancies caused by dysfunctional stem cells, leading to abnormal hematopoiesis and cytopenia. Approximately 30% of MDS cases progress to acute myeloid leukemia (AML), a more aggressive disease. Early detection is crucial to intervene before MDS progresses to AML. Artificial intelligence (AI) involves computer programs that can think like humans, and machine learning (ML) is a part of AI that helps computers learn patterns and make predictions. By using these technologies, doctors can improve how they diagnose MDS, leading to better treatment and outcomes for patients.

myelodysplastic syndrome diagnosis artificial intelligence machine learning

1. Introduction

Myelodysplastic syndrome (MDS) is a diverse group of hematological malignancies characterized by dysfunctional pluripotent stem cells that fail to undergo proper hematopoiesis and maturation within the bone marrow. Consequently, this leads to an excessive production of immature cells and dysplastic changes in the bone marrow. This disruption in stem cell activity results in a reduction in the formation of healthy blood cells, which manifests as cytopenia in one or more cell types, such as thrombocytopenia, erythrocytopenia, or leukocytopenia [1]. While the majority of adult MDS cases have no known etiology (primary or idiopathic), a small percentage of cases might be linked to an underlying illness (secondary), some of which are linked to autoinflammatory conditions termed VEXA syndrome [2][3]. This illness predominantly affects the elderly and usually has a gradual clinical course [4]. Patients’ presentation typically depends on the manifested cytopenia. They may develop anemia-related symptoms such as fatigue, weakness, and pallor. Recurrent infections and petechial bleeding may also develop as a result of a low number of functional leukocytes and platelets [5][6][7][8]. To establish a diagnosis of MDS, blood tests, a bone marrow biopsy, and genetic analysis are necessary. The diagnosis of MDS requires persistent cytopenia that cannot be explained by any other drug or cause, the presence of < 20% blasts on peripheral blood (PB) or bone marrow biopsy (BM) along with cytogenetic/molecular features (such as mutated SF3B1), or the presence of dysplastic morphology greater than 10% in a specific hematopoietic lineage without another explainable cause [9].
It is important to note that approximately 30% of patients with MDS will eventually develop acute myeloid leukemia (AML), which is more aggressive [10]. Hence, early diagnosis and treatment of MDS are crucial to improving patient outcomes [11]. MDS is a complex medical condition that can benefit from advancements in artificial intelligence (AI) and machine learning (ML). AI refers to the development of computer programs that emulate human intelligence. In healthcare, AI has the potential to improve the diagnosis, early detection, prognostication, and monitoring of diseases. Machine learning, a subset of AI, plays a crucial role in harnessing the power of datasets to recognize patterns and generate predictions. What sets ML algorithms apart is their ability to analyze both linear and nonlinear variables simultaneously, enabling them to identify complex patterns and make highly accurate predictions [12][13][14][15]. With the integration of AI and ML, healthcare providers can enhance the accuracy and efficiency of diagnosing MDS. The early diagnosis of MDS can lead to more informed decisions and early treatment plans for patients, leading to improved outcomes and better patient care.

2. Integrating AI and ML in Myelodysplastic Syndrome Diagnosis

2.1. Diagnosis of MDS Using BM Samples

BM smears are considered a prerequisite for the diagnosis of MDS. They provide a comprehensive view of cellular composition, morphology, and cytogenetics. The hallmarks of MDS on BM smears include dysplasia and elevated blasts that are <20%. The diagnosis of MDS with dysplasia is only possible when dysplasia reaches 10% in at least one lineage [16]. However, the analysis of BM samples for dysplasia and blasts, along with their quantification, can be difficult and time-consuming for pathologists, which can occasionally lead to the oversight of critical findings. Moreover, the assessment of dysplasia is subjective. Operators have to undergo years of training in order to become competent in the assessment of BM samples, and even then, inter- and intra-variations are present amongst experienced hematologists [17][18][19]. Herein lies the potential for AI to revolutionize MDS diagnosis. By harnessing AI’s capacity for rapid pattern recognition and data analysis, many challenges posed by manual examination of bone marrow samples can be mitigated.
To address the issue of identifying dysplasia, Lee and colleagues presented a convolutional neural network (CNN)-derived ML model that automatically detects dysplasia from images of bone marrow aspirates. The investigators acquired BM aspirates from 34 patients diagnosed with MDS and 24 patients without MDS. They manually captured images within well-spread areas containing nucleated cells to use as examples for the program. In order for the model to function, it had to be able to identify the cells and then classify them. For this, the researchers labeled the boundaries of 946 cells and classified 8065 cells into eight types (normal erythrocytes, normal granulocytes, normal megakaryocytes, dysplastic erythrocytes, dysplastic granulocytes, dysplastic megakaryocytes, blasts, and others). This was used to help train the model to identify and classify these cell types. 
The model proposed by Lee and colleagues demonstrated excellent ability in identifying the presence of dysplastic cells in three different lineages, but it is important to note that this model is not able to quantify the percentage of dysplasia. This makes it an excellent auxiliary tool to assist hematologists in recognizing dysplasia when attempting to diagnose MDS. Although the model by Lee was validated by competing with hematologists, it was not externally validated. This form of verification only provides insight into the quality of the model’s prediction and not its generalizability to other samples. Moreover, the model was not trained to distinguish specific changes in different cell types within the BM. 
Another model for the detection of dysplasia was proposed by Mori, J. et al. [20] Similar to the one by Lee, the model utilized images of BM smears from MDS and non-MDS patients with labeling performed by morphologists to assist the training of the model. However, Mori’s model utilized decreased granules (DGs) as a marker of dysplasia in granulopoiesis. They classified dysplasia on a 4-point scale, with 0 being normal, 1 intermediate, 2 dysplasia, and 3 severe dysplasia (i.e., severely decreased granules). 
The notable distinction of the model presented by Mori, J. et al. is that it relies on cellular features (granules) to detect dysplasia, unlike the model presented by Lee and colleagues. Moreover, the model classifies dysplasia by severity, not just dysplastic vs. non-dysplastic, which can be clinically useful. However, this model was neither externally validated nor challenged by hematologists. 
To address the issue of detecting blasts and quantifying them, Wu, Y. and colleagues presented an AI model that can detect and quantify blasts. BM smears were taken from patients with various hematological conditions. They were divided into a training sample (42), a testing sample (70), and a competition sample (10). Over seventeen thousand images of cells captured by hematologists from the training set were labeled and classified into one of seven cell categories (erythroid, blasts, myeloid, lymphoid, plasma cells, monocyte, and megakaryocyte) by three independent hematologists. If three hematologists could not agree on a cell’s type, they marked it as “unable to identify”. This information was used to train the CNN model to identify and categorize these cell types. To evaluate how well the CNN model performed compared to hematologists, a human-machine competition was conducted involving six visiting staff members. These staff members analyzed the same 10 BM samples from the competition cohort as the model. The results obtained from flow cytometry (FCM) were considered the established and accurate reference for comparison.
Another issue that pertains to the diagnosis of MDS is its differentiation from AA and leukemia. It is important to rule out AA when diagnosing MDS because these two conditions share some similar clinical and hematological features, especially hypocellular MDS [21][22]. Since both hematological conditions result in cytopenia, they can sometimes be confused for one another. Current diagnostic methods include hematologic analysis, bone marrow biopsy, cytogenetics, and flow cytometry (FC). Pathological hematopoiesis is nonspecific and occurs in both states. Once thought to be dependable, cytogenetic abnormalities are no longer reliably unique to MDS. While FC has grown in popularity, its single marker usage and limitations in detecting erythroid malignancies make it difficult to diagnose MDS in general [23][24][25]. 
To address this issue, a study by Wang et al. presented a deep learning model for the automatic diagnosis of MDS and the distinction between AA and AML based on BM smears [26]. The model was developed using a CNN and trained with data extracted from the American Society of Hematology (ASH) Image Bank, while external validation was performed using data from the clinic. Data from the ASH were randomly divided in a 7:3 ratio into training and testing datasets. Three different epochs were used for each model (30, 50, and 200). This determines the number of times the training set is presented to the learning model. The model had two output layers: whether the patient has MDS or not (two classifications) and whether they have AA, MDS, or AML (three classifications). The best model training effect was achieved with an outcome weight and epoch of 1:9 and 200, respectively. On external validation, the model exhibited high performance metrics in distinguishing MDS from non-MDS (AUC: 0.942, ACC: 0.921, SEN: 0.886, SPE: 0.938) and in distinguishing MDS, AA, and AML (AUC: 0.948, ACC: 0.915, SEN: 0.887, SPE: 0.929) [26]. Overall, the image-net pretrained model provided a convenient and accurate tool for clinicians to differentiate AA, MDS, and AML based on bone marrow smear images.

2.2. Diagnosis of MDS Using PBS

The conventional diagnosis of MDS from peripheral blood smears (PBS) presents its own set of challenges. PBS offer a snapshot of hematological abnormalities and can provide crucial insights into the diagnosis of MDS. However, similar to BM smears, the manual examination of PBS is time-consuming, subject to human error, and often requires experienced hematologists [27]. These challenges have paved the way for the application of AI techniques to enhance the accuracy, efficiency, and objectivity of MDS diagnosis using peripheral blood smears.
Multiple studies have shown that hypogranulated dysplastic neutrophils on PBS can provide valuable insights into the diagnosis of MDS [28][29][30][31]. However, it is sometimes challenging for pathologists to identify them on PBS. Hence, Acevedo and colleagues aimed to address the issue of identifying hypogranulated dysplastic neutrophils in peripheral blood by developing eight ML models labeled M1 to M8 using a CNN to undertake this task [20]. These models varied in architectural elements and training methodologies but were all trained for 20 epochs.
Another model was also proposed by Kimura et al. for automatic MDS differentiation from AA through a CNN utilizing PBS data [32]. They combined a CNN-powered DLS with the automatic detection and recognition of blood cells with an XGBoost decision-making system. Over 690,000 blood cell images from 3281 PBS were utilized in the training of their CNN model. Their model was able to classify 17 different blood cell types and their 97 morphological characteristics with an impressive SEN and SPE of 0.935 and 0.960, respectively. Their final model was able to distinguish MDS from AA utilizing PBS data with an AUC of 0.99, SEN of 0.962, SPE of 1.00, and overall ACC of 0.900. The limitations of their model included the adjunctive nature of the system, requiring additional diagnostic methods, and the need for clinical and genetic data for a definitive diagnosis. The study acknowledged the small sample size and single-center design, proposing future work to expand the dataset and enhance accuracy using serum biochemistry data [32].

2.3. Diagnosis of MDS Using FC

FC serves as a crucial tool in the diagnosis of MDS, aiding in the recognition of specific cellular attributes and counts that characterize this complex hematologic disorder. By enabling the precise analysis of individual cells, FC assists in identifying distinct markers and aberrant expression patterns that are indicative of MDS [33][34][35]. Despite its utility, the current utilization of FC faces challenges such as labor-intensive manual data interpretation, subjectivity in gating procedures, and a lack of standardized quantification, all of which hinder its efficiency and consistency in MDS diagnosis [33][36]. To overcome these limitations, AI emerges as a potential solution. AI offers the capacity to automate and optimize the analysis of high-dimensional flow cytometry data using advanced machine learning techniques. AI has the potential to enhance diagnostic accuracy, reduce variability, and uncover subtle cellular features that may hold diagnostic significance. Integrating AI into flow cytometry-based MDS diagnosis has the potential to revolutionize the field, addressing current limitations and providing a more efficient and precise approach to characterizing this challenging hematologic disorder.
Valentin Clichet et al. introduced an innovative approach combining AI with multiparametric FC to enhance MDS diagnosis and classification [37]. Their machine learning model employed an elasticnet algorithm applied to a cohort of 191 patients suspected of MDS. The research focused solely on flow cytometry parameters and utilized the Boruta algorithm for feature selection in the model. Granulocyte/lymphocyte SSC peak channel ratio, total hematogone ratio, percentage of CD34+ B-cell progenitors among all CD34+ cells, and the percentage of CD34+ myeloid progenitors were found to be the most important predictors for MDS diagnosis by the Boruta algorithm. The AI-assisted MDS prediction score (elasticnet model) demonstrates superior sensitivity to the existing Ogata score, maintaining excellent specificity. An external validation cohort of 89 patients confirms its high performance, with an AUC of 0.935. Notably, this model effectively diagnoses both high- and low-risk MDS, achieving 91.8% SEN and 92.5% SPE. Moreover, it reveals a progressive evolution of the prediction score from clonal hematopoiesis of indeterminate potential (CHIP) to high-risk MDS, implying a linear progression between these stages.
The significance of the research lies not only in its accuracy but also in its insights into disease classification. The authors emphasize that even with only half of the FC markers, the algorithm maintains high recognition accuracy, shedding light on the discriminability of existing markers. Moreover, the approach highlights the potential for reducing marker redundancy through computational methods. This novel algorithm consistently outperforms other representations across various classification tasks, emphasizing the importance of cell-level feature representation facilitated by autoencoder learning. While the findings of the study hold promise for advancing MRD classification, certain limitations warrant consideration. The observed discrepancies in classification accuracy between AML, MDS, and normal categories might stem from inherent complexities in categorizing MDS and potential data imbalances. Furthermore, the study’s focus on a specific dataset and markers necessitates further exploration to validate its applicability across broader contexts.

3. Conclusions

In conclusion, while the utilization of machine learning algorithms holds significant promise in the diagnosis of MDS, the current landscape is characterized by a limited yet encouraging body of research. These studies, employing various datasets, including PBS, BMS, and FC data, have exhibited noteworthy potential for accurately diagnosing and stratifying MDS patients. However, the absence of comprehensive external validation, coupled with the need for integrating diverse data sources representative of the multimodal diagnostic approach, underscores the imperative for cautious optimism. As AI continues its transformative journey in hematological disease diagnosis, its role as an assisting tool for pathologists and hematologists remains a compelling avenue, warranting further investigation and validation to unlock its full clinical potential in MDS management.

References

  1. Ades, L.; Itzykson, R.; Fenaux, P. Myelodysplastic syndromes. Lancet 2014, 383, 2239–2252.
  2. Muslimani, A.A.; Spiro, T.P.; Chaudhry, A.A.; Daw, H.A. Secondary myelodysplastic syndrome after hydroxychloroquine therapy. Ann. Hematol. 2007, 86, 531–534.
  3. Beck, D.B.; Ferrada, M.A.; Sikora, K.A.; Ombrello, A.K.; Collins, J.C.; Pei, W.; Balanda, N.; Ross, D.L.; Ospina Cardona, D.; Wu, Z.; et al. Somatic Mutations in UBA1 and Severe Adult-Onset Autoinflammatory Disease. N. Engl. J. Med. 2020, 383, 2628–2638.
  4. Zeidan, A.M.; Shallis, R.M.; Wang, R.; Davidoff, A.; Ma, X. Epidemiology of myelodysplastic syndromes: Why characterizing the beast is a prerequisite to taming it. Blood Rev. 2019, 34, 1–15.
  5. Goldberg, S.L.; Chen, E.; Corral, M.; Guo, A.; Mody-Patel, N.; Pecora, A.L.; Laouri, M. Incidence and clinical complications of myelodysplastic syndromes among United States Medicare beneficiaries. J. Clin. Oncol. 2010, 28, 2847–2852.
  6. Meyers, C.A.; Albitar, M.; Estey, E. Cognitive impairment, fatigue, and cytokine levels in patients with acute myelogenous leukemia or myelodysplastic syndrome. Cancer 2005, 104, 788–793.
  7. Sekeres, M.A.; Taylor, J. Diagnosis and Treatment of Myelodysplastic Syndromes: A Review. JAMA 2022, 328, 872–880.
  8. Al-Haidose, A.; Yassin, M.A.; Ahmed, M.N.; Kunhipurayil, H.H.; Al-Harbi, A.A.; Aljaberi, M.A.; Abbasi, S.A.; Kordasti, S.; Abdallah, A.M. Distinct Clinical and Prognostic Features of Myelodysplastic Syndrome in Patients from the Middle East, North Africa, and Beyond: A Systemic Review. J. Clin. Med. 2023, 12, 2832.
  9. Khoury, J.D.; Solary, E.; Abla, O.; Akkari, Y.; Alaggio, R.; Apperley, J.F.; Bejar, R.; Berti, E.; Busque, L.; Chan, J.K.C.; et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Myeloid and Histiocytic/Dendritic Neoplasms. Leukemia 2022, 36, 1703–1719.
  10. Estey, E.; Hasserjian, R.P.; Dohner, H. Distinguishing AML from MDS: A fixed blast percentage may no longer be optimal. Blood 2022, 139, 323–332.
  11. Steensma, D.P. Does early diagnosis and treatment of myelodysplastic syndromes make a difference? Best Pract. Res. Clin. Haematol. 2019, 32, 101099.
  12. Al-Antari, M.A. Artificial Intelligence for Medical Diagnostics-Existing and Future AI Technology! Diagnostics 2023, 13, 688.
  13. Davenport, T.; Kalakota, R. The potential for artificial intelligence in healthcare. Future Healthc. J. 2019, 6, 94–98.
  14. Kumar, Y.; Koul, A.; Singla, R.; Ijaz, M.F. Artificial intelligence in disease diagnosis: A systematic literature review, synthesizing framework and future research agenda. J. Ambient. Intell. Humaniz. Comput. 2023, 14, 8459–8486.
  15. Undru, T.R.; Uday, U.; Lakshmi, J.T.; Kaliappan, A.; Mallamgunta, S.; Nikhat, S.S.; Sakthivadivel, V.; Gaur, A. Integrating Artificial Intelligence for Clinical and Laboratory Diagnosis—A Review. Maedica 2022, 17, 420–426.
  16. Dao, K.T. Myelodysplastic Syndromes: Updates and Nuances. Med. Clin. N. Am. 2017, 101, 333–350.
  17. Mantripragada, V.P.; Piuzzi, N.S.; George, J.; Bova, W.; Ng, M.; Boehm, C.; Muschler, G.F. Reliable assessment of bone marrow and bone marrow concentrates using automated hematology analyzer. Regen. Med. 2019, 14, 639–646.
  18. Percival, M.E.; Lai, C.; Estey, E.; Hourigan, C.S. Bone marrow evaluation for diagnosis and monitoring of acute myeloid leukemia. Blood Rev. 2017, 31, 185–192.
  19. Piuzzi, N.S.; Hussain, Z.B.; Chahla, J.; Cinque, M.E.; Moatshe, G.; Mantripragada, V.P.; Muschler, G.F.; LaPrade, R.F. Variability in the Preparation, Reporting, and Use of Bone Marrow Aspirate Concentrate in Musculoskeletal Disorders: A Systematic Review of the Clinical Orthopaedic Literature. J. Bone Jt. Surg. Am. 2018, 100, 517–525.
  20. Acevedo, A.; Merino, A.; Boldu, L.; Molina, A.; Alferez, S.; Rodellar, J. A new convolutional neural network predictive model for the automatic recognition of hypogranulated neutrophils in myelodysplastic syndromes. Comput. Biol. Med. 2021, 134, 104479.
  21. Barrett, J.; Saunthararajah, Y.; Molldrem, J. Myelodysplastic syndrome and aplastic anemia: Distinct entities or diseases linked by a common pathophysiology? Semin. Hematol. 2000, 37, 15–29.
  22. DeZern, A.E.; Churpek, J.E. Approach to the diagnosis of aplastic anemia. Blood Adv. 2021, 5, 2660–2671.
  23. DeZern, A.E.; Sekeres, M.A. The challenging world of cytopenias: Distinguishing myelodysplastic syndromes from other disorders of marrow failure. Oncologist 2014, 19, 735–745.
  24. Durrani, J.; Maciejewski, J.P. Idiopathic aplastic anemia vs hypocellular myelodysplastic syndrome. Hematol. Am. Soc. Hematol. Educ. Program. 2019, 2019, 97–104.
  25. Keel, S.B.; Scott, A.; Sanchez-Bonilla, M.; Ho, P.A.; Gulsuner, S.; Pritchard, C.C.; Abkowitz, J.L.; King, M.C.; Walsh, T.; Shimamura, A. Genetic features of myelodysplastic syndrome and aplastic anemia in pediatric and young adult patients. Haematologica 2016, 101, 1343–1350.
  26. Wang, M.; Dong, C.; Gao, Y.; Li, J.; Han, M.; Wang, L. A Deep Learning Model for the Automatic Recognition of Aplastic Anemia, Myelodysplastic Syndromes, and Acute Myeloid Leukemia Based on Bone Marrow Smear. Front. Oncol. 2022, 12, 844978.
  27. Mohammed, E.A.; Mohamed, M.M.; Far, B.H.; Naugler, C. Peripheral blood smear image analysis: A comprehensive review. J. Pathol. Inform. 2014, 5, 9.
  28. Gupta, G.; Singh, R.; Kotasthane, D.S.; Kotasthane, V.D. Myelodysplastic syndromes/neoplasms: Recent classification system based on World Health Organization Classification of Tumors—International Agency for Research on Cancer for Hematopoietic and Lymphoid Tissues. J. Blood Med. 2010, 1, 171–182.
  29. Hast, R.; Nilsson, I.; Widell, S.; Ost, A. Diagnostic significance of dysplastic features of peripheral blood polymorphs in myelodysplastic syndromes. Leuk. Res. 1989, 13, 173–178.
  30. Parmentier, S.; Schetelig, J.; Lorenz, K.; Kramer, M.; Ireland, R.; Schuler, U.; Ordemann, R.; Rall, G.; Schaich, M.; Bornhauser, M.; et al. Assessment of dysplastic hematopoiesis: Lessons from healthy bone marrow donors. Haematologica 2012, 97, 723–730.
  31. Widell, S.; Hellstrom-Lindberg, E.; Kock, Y.; Lindberg, M.; Ost, A.; Hast, R. Peripheral blood neutrophil morphology reflects bone marrow dysplasia in myelodysplastic syndromes. Am. J. Hematol. 1995, 49, 115–120.
  32. Kimura, K.; Tabe, Y.; Ai, T.; Takehara, I.; Fukuda, H.; Takahashi, H.; Naito, T.; Komatsu, N.; Uchihashi, K.; Ohsaka, A. A novel automated image analysis system using deep convolutional neural networks can assist to differentiate MDS and AA. Sci. Rep. 2019, 9, 13385.
  33. Bento, L.C.; Correia, R.P.; Pitangueiras Mangueira, C.L.; De Souza Barroso, R.; Rocha, F.A.; Bacal, N.S.; Marti, L.C. The Use of Flow Cytometry in Myelodysplastic Syndromes: A Review. Front. Oncol. 2017, 7, 270.
  34. Oelschlaegel, U.; Oelschlaeger, L.; von Bonin, M.; Kramer, M.; Sockel, K.; Mohr, B.; Wagenfuehr, L.; Kroschinsky, F.; Bornhaeuser, M.; Platzbecker, U. Comparison of five diagnostic flow cytometry scores in patients with myelodysplastic syndromes: Diagnostic power and prognostic impact. Cytom. B Clin. Cytom. 2023, 104, 141–150.
  35. Pembroke, J.S.; Joseph, J.E.; Smith, S.; Parker, A.J.C.; Jiang, W.; Sewell, W.A. Comparison of flow cytometry with other modalities in the diagnosis of myelodysplastic syndrome. Int. J. Lab. Hematol. 2022, 44, 313–319.
  36. van de Loosdrecht, A.A.; Alhan, C.; Bene, M.C.; Della Porta, M.G.; Drager, A.M.; Feuillard, J.; Font, P.; Germing, U.; Haase, D.; Homburg, C.H.; et al. Standardization of flow cytometry in myelodysplastic syndromes: Report from the first European LeukemiaNet working conference on flow cytometry in myelodysplastic syndromes. Haematologica 2009, 94, 1124–1134.
  37. Clichet, V.; Lebon, D.; Chapuis, N.; Zhu, J.; Bardet, V.; Marolleau, J.P.; Garcon, L.; Caulier, A.; Boyer, T. Artificial intelligence to empower diagnosis of myelodysplastic syndromes by multiparametric flow cytometry. Haematologica 2023, 108, 2435–2443.
More
© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license; additional terms may apply. By using this site, you agree to the Terms and Conditions and Privacy Policy.
Upload a video for this entry
Information
Subjects: Hematology
Contributors MDPI registered users' name will be linked to their SciProfiles pages. To register with us, please refer to https://encyclopedia.pub/register :
Amgad Mohamed Elshoeibi
,
Ahmed Badr
,
Basel Elsayed
,
Omar Metwally
,
Raghad Elshoeibi
,
Mohamed Ragab Elhadary
,
Ahmed Elshoeibi
,
Mohamed Amro Attya
,
Fatima Khadadah
,
Awni Alshurafa
,
Ahmad Alhuraiji
,
Mohamed Yassin
View Times: 85
Revisions: 2 times (View History)
Update Date: 21 Feb 2024
Table of Contents
    1000/1000
    Hot Most Recent
    Feedback