| 1. |
|
|
| 2. |
- Zhou, Zheng, et al.
(författare)
-
Reduced intensity conditioning for acute myeloid leukemia using melphalan- vs busulfan-based regimens : a CIBMTR report
- 2020
-
Ingår i: Blood Advances. - : American Society of Hematology. - 2473-9529 .- 2473-9537. ; 4:13, s. 3180-3190
-
Tidskriftsartikel (refereegranskat)abstract
- There is a lack of large comparative study on the outcomes of reduced intensity conditioning (RIC) in acute myeloid leukemia (AML) transplantation using fludarabine/busulfan (FB) and fludarabine/melphalan (FM) regimens. Adult AML patients from Center for International Blood and Marrow Transplant Research who received first RIC allo-transplant between 2001 and 2015 were studied. Patients were excluded if they received cord blood or identical twin transplant, total body irradiation in conditioning, or graft-versus-host disease (GVHD) prophylaxis with in vitro T-cell depletion. Primary outcome was overall survival (OS), secondary end points were leukemia-free survival (LFS), nonrelapse mortality (NRM), relapse, and GVHD. Multivariate survival model was used with adjustment for patient, leukemia, and transplant-related factors. A total of 622 patients received FM and 791 received FB RIC. Compared with FB, the FM group had fewer transplant in complete remission (CR), fewer matched sibling donors, and less usage of anti-thymocyte globulin or alemtuzumab. More patients in the FM group received marrow grafts and had transplantation before 2005. OS was significantly lower within the first 3 months posttransplant in the FM group (hazard ratio [HR] = 1.82, P < .001), but was marginally superior beyond 3 months (HR = 0.87, P = .05). LFS was better with FM compared with FB (HR = 0.89, P = .05). NRM was significantly increased in the FM group during the first 3 months of posttransplant (HR = 3.85, P < .001). Long-term relapse was lower with FM (HR = 0.65, P < .001). Analysis restricted to patients with CR showed comparable results. In conclusion, compared with FB, the FM RIC showed a marginally superior long-term OS and LFS and a lower relapse rate. A lower OS early posttransplant within 3 months was largely the result of a higher early NRM.
|
|
| 3. |
- DeFilipp, Zachariah, et al.
(författare)
-
Maintenance Tyrosine Kinase Inhibitors Following Allogeneic Hematopoietic Stem Cell Transplantation for Chronic Myelogenous Leukemia : A Center for International Blood and Marrow Transplant Research Study
- 2020
-
Ingår i: Biology of blood and marrow transplantation. - : Elsevier. - 1083-8791 .- 1523-6536. ; 26:3, s. 472-479
-
Tidskriftsartikel (refereegranskat)abstract
- It remains unknown whether the administration of tyrosine kinase inhibitors (TKIs) targeting BCR-ABL1 after allogeneic hematopoietic cell transplantation (HCT) is associated with improved outcomes for patients with chronic myelogenous leukemia (CML). In this registry study, we analyzed clinical outcomes of 390 adult patients with CML who underwent transplantation between 2007 and 2014 and received maintenance TKI following HCT (n = 89) compared with no TKI maintenance (n = 301), as reported to the Center for International Blood and Marrow Transplant Research. All patients received TKI therapy before HCT. The majority of patients had a disease status of first chronic phase at HCT (n = 240; 62%). The study was conducted as a landmark analysis, excluding patients who died, relapsed, had chronic graft-versus-host disease, or were censored before day +100 following HCT. Of the 89 patients who received TKI maintenance, 77 (87%) received a single TKI and the other 12 (13%) received multiple sequential TKIs. The most common TKIs used for maintenance were dasatinib (n = 50), imatinib (n = 27), and nilotinib (n = 27). As measured from day +100, the adjusted estimates for 5-year relapse (maintenance, 35% versus no maintenance, 26%; P = .11), leukemia-free survival (maintenance, 42% versus no maintenance, 44%; P = .65), or overall survival (maintenance, 61% versus no maintenance, 57%; P = .61) did not differ significantly between patients receiving TKI maintenance or no maintenance. These results remained unchanged in multivariate analysis and were not modified by disease status before transplantation. In conclusion, our data from this day +100 landmark analysis do not demonstrate a significant impact of maintenance TKI therapy on clinical outcomes. The optimal approach to TKI administration in the post-transplantation setting in patients with CML remains undetermined.
|
|
| 4. |
- Horta Ribeiro, Antônio, et al.
(författare)
-
Regularization properties of adversarially-trained linear regression
- 2023
-
Ingår i: Advances in Neural Information Processing Systems 36 (NeurIPS 2023). - : Curran Associates, Inc.. ; , s. 23658-23670
-
Konferensbidrag (refereegranskat)abstract
- State-of-the-art machine learning models can be vulnerable to very small input perturbations that are adversarially constructed. Adversarial training is an effective approach to defend against it. Formulated as a min-max problem, it searches for the best solution when the training data were corrupted by the worst-case attacks. Linear models are among the simple models where vulnerabilities can be observed and are the focus of our study. In this case, adversarial training leads to a convex optimization problem which can be formulated as the minimization of a finite sum. We provide a comparative analysis between the solution of adversarial training in linear regression and other regularization methods. Our main findings are that: (A) Adversarial training yields the minimum-norm interpolating solution in the overparameterized regime (more parameters than data), as long as the maximum disturbance radius is smaller than a threshold. And, conversely, the minimum-norm interpolator is the solution to adversarial training with a given radius. (B) Adversarial training can be equivalent to parameter shrinking methods (ridge regression and Lasso). This happens in the underparametrized region, for an appropriate choice of adversarial radius and zero-mean symmetrically distributed covariates. (C) For l(infinity)-adversarial training-as in square-root Lasso-the choice of adversarial radius for optimal bounds does not depend on the additive noise variance. We confirm our theoretical findings with numerical examples.
|
|
| 5. |
|
|
| 6. |
- Osama, Muhammad, et al.
(författare)
-
Online Learning for Prediction via Covariance Fitting : Computation, Performance and Robustness
- 2023
-
Ingår i: Transactions on Machine Learning Research. - : Transactions on Machine Learning Research. - 2835-8856.
-
Tidskriftsartikel (refereegranskat)abstract
- We consider the online learning of linear smoother predictors based on a covariance model of the outcomes. To control its degrees of freedom in an appropriate manner, the covariance model parameters are often learned using cross-validation or maximum-likelihood techniques. However, neither technique is suitable when training data arrives in a streaming fashion. Here we consider a covariance-fitting method to learn the model parameters, initially used in spectral estimation. We show that this results in a computation efficient online learning method in which the resulting predictor can be updated sequentially. We prove that, with high probability, its out-of-sample error approaches the minimum achievable level at root-$n$ rate. Moreover, we show that the resulting predictor enjoys two different robustness properties. First, it minimizes the out-of-sample error with respect to the least favourable distribution within a given Wasserstein distance from the empirical distribution. Second, it is robust against errors in the covariate training data. We illustrate the performance of the proposed method in a numerical experiment.
|
|
| 7. |
- Osama, Muhammad
(författare)
-
Robust machine learning methods
- 2022
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- We are surrounded by data in our daily lives. The rent of our houses, the amount of electricity units consumed, the prices of different products at a supermarket, the daily temperature, our medicine prescriptions, our internet search history are all different forms of data. Data can be used in a wide range of applications. For example, one can use data to predict product prices in the future; to predict tomorrow's temperature; to recommend videos; or suggest better prescriptions. However in order to do the above, one is required to learn a model from data. A model is a mathematical description of how the phenomena we are interested in behaves e.g. how does the temperature vary? Is it periodic? What kinds of patterns does it have? Machine learning is about this process of learning models from data by building on disciplines such as statistics and optimization. Learning models comes with many different challenges. Some challenges are related to how flexible the model is, some are related to the size of data, some are related to computational efficiency etc. One of the challenges is that of data outliers. For instance, due to war in a country exports could stop and there could be a sudden spike in prices of different products. This sudden jump in prices is an outlier or corruption to the normal situation and must be accounted for when learning the model. Another challenge could be that data is collected in one situation but the model is to be used in another situation. For example, one might have data on vaccine trials where the participants were mostly old people. But one might want to make a decision on whether to use the vaccine or not for the whole population that contains people of all age groups. So one must also account for this difference when learning models because the conclusion drawn may not be valid for the young people in the population. Yet another challenge could arise when data is collected from different sources or contexts. For example, a shopkeeper might have data on sales of paracetamol when there was flu and when there was no flu and she might want to decide how much paracetamol to stock for the next month. In this situation, it is difficult to know whether there will be a flu next month or not and so deciding on how much to stock is a challenge. This thesis tries to address these and other similar challenges.In paper I, we address the challenge of data corruption i.e., learning models in a robust way when some fraction of the data is corrupted. In paper II, we apply the methodology of paper I to the problem of localization in wireless networks. Paper III addresses the challenge of estimating causal effect between an exposure and an outcome variable from spatially collected data (e.g. whether increasing number of police personnel in an area reduces number of crimes there). Paper IV addresses the challenge of learning improved decision policies e.g. which treatment to assign to which patient given past data on treatment assignments. In paper V, we look at the challenge of learning models when data is acquired from different contexts and the future context is unknown. In paper VI, we address the challenge of predicting count data across space e.g. number of crimes in an area and quantify its uncertainty. In paper VII, we address the challenge of learning models when data points arrive in a streaming fashion i.e., point by point. The proposed method enables online training and also yields some robustness properties.
|
|
| 8. |
- Percival, Mary-Elizabeth, et al.
(författare)
-
Impact of depth of clinical response on outcomes of acute myeloid leukemia patients in first complete remission who undergo allogeneic hematopoietic cell transplantation
- 2021
-
Ingår i: Bone Marrow Transplantation. - : Springer Nature. - 0268-3369 .- 1476-5365. ; 56:9, s. 2108-2117
-
Tidskriftsartikel (refereegranskat)abstract
- Acute myeloid leukemia (AML) patients often undergo allogeneic hematopoietic cell transplantation (alloHCT) in first complete remission (CR). We examined the effect of depth of clinical response, including incomplete count recovery (CRi) and/or measurable residual disease (MRD), in patients from the Center for International Blood and Marrow Transplantation Research (CIBMTR) registry. We identified 2492 adult patients (1799 CR and 693 CRi) who underwent alloHCT between January 1, 2007 and December 31, 2015. The primary outcome was overall survival (OS). Multivariable analysis was performed to adjust for patient-, disease-, and transplant-related factors. Baseline characteristics were similar. Patients in CRi compared to those in CR had an increased likelihood of death (HR: 1.27; 95% confidence interval: 1.13-1.43). Compared to CR, CRi was significantly associated with increased non-relapse mortality (NRM), shorter disease-free survival (DFS), and a trend toward increased relapse. Detectable MRD was associated with shorter OS, shorter DFS, higher NRM, and increased relapse compared to absence of MRD. The deleterious effects of CRi and MRD were independent. In this large CIBMTR cohort, survival outcomes differ among AML patients based on depth of CR and presence of MRD at the time of alloHCT. Further studies should focus on optimizing post-alloHCT outcomes for patients with responses less than CR.
|
|
