Search

Research Question: Can federated learning be used to develop an artificial intelligence (AI) model for evaluating oocyte competence using two-dimensional images of denuded oocytes in metaphase II prior to intracytoplasmic sperm injection (ICSI)?, Results: The oocyte AI model demonstrated area under the curve (AUC) up to 0.65 on two blind test datasets. High sensitivity for predicting competent oocytes (83-88%) was offset by lower specificity (26-36%). Exclusion of confounding biological variables (male factor infertility and maternal age ≥35 years) improved AUC up to 14%, primarily due to increased specificity. AI score correlated with size of the zona pellucida and perivitelline space, and ooplasm appearance. AI score also correlated with blastocyst expansion grade and morphological quality. The sum of AI scores from oocytes in group culture images predicted the formation of two or more usable blastocysts (AUC 0.77)., Conclusion: An AI model to evaluate oocyte competence was developed using federated learning, representing an essential step in protecting patient data. The AI model was significantly predictive of oocyte competence, as defined by usable blastocyst formation, which is a critical factor for IVF success. Potential clinical utility ranges from selective oocyte fertilization to guiding treatment decisions regarding additional rounds of oocyte retrieval., Design: In total, 10,677 oocyte images with associated metadata were collected prospectively by eight IVF clinics across six countries. AI training used federated learning, where data were retained on regional servers to comply with data privacy laws. The final AI model required a single image as input to evaluate oocyte competence, which was defined by the formation of a usable blastocyst (≥expansion grade 3 by day 5 or 6 post ICSI)., (Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Medical datasets inherently contain errors from subjective or inaccurate test results, or from confounding biological complexities. It is difficult for medical experts to detect these elusive errors manually, due to lack of contextual information, limiting data privacy regulations, and the sheer scale of data to be reviewed. Current methods for training robust artificial intelligence (AI) models on data containing mislabeled examples generally fall into one of several categories-attempting to improve the robustness of the model architecture, the regularization techniques used, the loss function used during training, or selecting a subset of data that contains cleaner labels. This last category requires the ability to efficiently detect errors either prior to or during training, either relabeling them or removing them completely. More recent progress in error detection has focused on using multi-network learning to minimize deleterious effects of errors on training, however, using many neural networks to reach a consensus on which data should be removed can be computationally intensive and inefficient. In this work, a deep-learning based algorithm was used in conjunction with a label-clustering approach to automate error detection. For dataset with synthetic label flips added, these errors were identified with an accuracy of up to 85%, while requiring up to 93% less computing resources to complete compared to a previous model consensus approach developed previously. The resulting trained AI models exhibited greater training stability and up to a 45% improvement in accuracy, from 69 to over 99% compared to the consensus approach, at least 10% improvement on using noise-robust loss functions in a binary classification problem, and a 51% improvement for multi-class classification. These results indicate that practical, automated a priori detection of errors in medical data is possible, without human oversight., (© 2023. The Author(s).)

Research Question: Can better methods be developed to evaluate the performance and characteristics of an artificial intelligence model for evaluating the likelihood of clinical pregnancy based on analysis of day-5 blastocyst-stage embryos, such that performance evaluation more closely reflects clinical use in IVF procedures, and correlations with known features of embryo quality are identified?, Design: De-identified images were provided retrospectively or collected prospectively by IVF clinics using the artificial intelligence model in clinical practice. A total of 9359 images were provided by 18 IVF clinics across six countries, from 4709 women who underwent IVF between 2011 and 2021. Main outcome measures included clinical pregnancy outcome (fetal heartbeat at first ultrasound scan), embryo morphology score, and/or pre-implantation genetic testing for aneuploidy (PGT-A) results., Results: A positive linear correlation of artificial intelligence scores with pregnancy outcomes was found, and up to a 12.2% reduction in time to pregnancy (TTP) was observed when comparing the artificial intelligence model with standard morphological grading methods using a novel simulated cohort ranking method. Artificial intelligence scores were significantly correlated with known morphological features of embryo quality based on the Gardner score, and with previously unknown morphological features associated with embryo ploidy status, including chromosomal abnormalities indicative of severity when considering embryos for transfer during IVF., Conclusion: Improved methods for evaluating artificial intelligence for embryo selection were developed, and advantages of the artificial intelligence model over current grading approaches were highlighted, strongly supporting the use of the artificial intelligence model in a clinical setting., (Copyright © 2022 The Author(s). Published by Elsevier Ltd.. All rights reserved.)

Study Question: Can an artificial intelligence (AI) model predict human embryo ploidy status using static images captured by optical light microscopy?, Summary Answer: Results demonstrated predictive accuracy for embryo euploidy and showed a significant correlation between AI score and euploidy rate, based on assessment of images of blastocysts at Day 5 after IVF., What Is Known Already: Euploid embryos displaying the normal human chromosomal complement of 46 chromosomes are preferentially selected for transfer over aneuploid embryos (abnormal complement), as they are associated with improved clinical outcomes. Currently, evaluation of embryo genetic status is most commonly performed by preimplantation genetic testing for aneuploidy (PGT-A), which involves embryo biopsy and genetic testing. The potential for embryo damage during biopsy, and the non-uniform nature of aneuploid cells in mosaic embryos, has prompted investigation of additional, non-invasive, whole embryo methods for evaluation of embryo genetic status., Study Design, Size, Duration: A total of 15 192 blastocyst-stage embryo images with associated clinical outcomes were provided by 10 different IVF clinics in the USA, India, Spain and Malaysia. The majority of data were retrospective, with two additional prospectively collected blind datasets provided by IVF clinics using the genetics AI model in clinical practice. Of these images, a total of 5050 images of embryos on Day 5 of in vitro culture were used for the development of the AI model. These Day 5 images were provided for 2438 consecutively treated women who had undergone IVF procedures in the USA between 2011 and 2020. The remaining images were used for evaluation of performance in different settings, or otherwise excluded for not matching the inclusion criteria., Participants/materials, Setting, Methods: The genetics AI model was trained using static 2-dimensional optical light microscope images of Day 5 blastocysts with linked genetic metadata obtained from PGT-A. The endpoint was ploidy status (euploid or aneuploid) based on PGT-A results. Predictive accuracy was determined by evaluating sensitivity (correct prediction of euploid), specificity (correct prediction of aneuploid) and overall accuracy. The Matthew correlation coefficient and receiver-operating characteristic curves and precision-recall curves (including AUC values), were also determined. Performance was also evaluated using correlation analyses and simulated cohort studies to evaluate ranking ability for euploid enrichment., Main Results and the Role of Chance: Overall accuracy for the prediction of euploidy on a blind test dataset was 65.3%, with a sensitivity of 74.6%. When the blind test dataset was cleansed of poor quality and mislabeled images, overall accuracy increased to 77.4%. This performance may be relevant to clinical situations where confounding factors, such as variability in PGT-A testing, have been accounted for. There was a significant positive correlation between AI score and the proportion of euploid embryos, with very high scoring embryos (9.0-10.0) twice as likely to be euploid than the lowest-scoring embryos (0.0-2.4). When using the genetics AI model to rank embryos in a cohort, the probability of the top-ranked embryo being euploid was 82.4%, which was 26.4% more effective than using random ranking, and ∼13-19% more effective than using the Gardner score. The probability increased to 97.0% when considering the likelihood of one of the top two ranked embryos being euploid, and the probability of both top two ranked embryos being euploid was 66.4%. Additional analyses showed that the AI model generalized well to different patient demographics and could also be used for the evaluation of Day 6 embryos and for images taken using multiple time-lapse systems. Results suggested that the AI model could potentially be used to differentiate mosaic embryos based on the level of mosaicism., Limitations, Reasons for Caution: While the current investigation was performed using both retrospectively and prospectively collected data, it will be important to continue to evaluate real-world use of the genetics AI model. The endpoint described was euploidy based on the clinical outcome of PGT-A results only, so predictive accuracy for genetic status in utero or at birth was not evaluated. Rebiopsy studies of embryos using a range of PGT-A methods indicated a degree of variability in PGT-A results, which must be considered when interpreting the performance of the AI model., Wider Implications of the Findings: These findings collectively support the use of this genetics AI model for the evaluation of embryo ploidy status in a clinical setting. Results can be used to aid in prioritizing and enriching for embryos that are likely to be euploid for multiple clinical purposes, including selection for transfer in the absence of alternative genetic testing methods, selection for cryopreservation for future use or selection for further confirmatory PGT-A testing, as required., Study Funding/competing Interest(s): Life Whisperer Diagnostics is a wholly owned subsidiary of the parent company, Presagen Holdings Pty Ltd. Funding for the study was provided by Presagen with grant funding received from the South Australian Government: Research, Commercialisation, and Startup Fund (RCSF). 'In kind' support and embryology expertise to guide algorithm development were provided by Ovation Fertility. 'In kind' support in terms of computational resources provided through the Amazon Web Services (AWS) Activate Program. J.M.M.H., D.P. and M.P. are co-owners of Life Whisperer and Presagen. S.M.D., M.A.D. and T.V.N. are employees or former employees of Life Whisperer. S.M.D, J.M.M.H, M.A.D, T.V.N., D.P. and M.P. are listed as inventors of patents relating to this work, and also have stock options in the parent company Presagen. M.V. sits on the advisory board for the global distributor of the technology described in this study and also received support for attending meetings., Trial Registration Number: N/A., (© The Author(s) 2022. Published by Oxford University Press on behalf of European Society of Human Reproduction and Embryology.)

Training on multiple diverse data sources is critical to ensure unbiased and generalizable AI. In healthcare, data privacy laws prohibit data from being moved outside the country of origin, preventing global medical datasets being centralized for AI training. Data-centric, cross-silo federated learning represents a pathway forward for training on distributed medical datasets. Existing approaches typically require updates to a training model to be transferred to a central server, potentially breaching data privacy laws unless the updates are sufficiently disguised or abstracted to prevent reconstruction of the dataset. Here we present a completely decentralized federated learning approach, using knowledge distillation, ensuring data privacy and protection. Each node operates independently without needing to access external data. AI accuracy using this approach is found to be comparable to centralized training, and when nodes comprise poor-quality data, which is common in healthcare, AI accuracy can exceed the performance of traditional centralized training., (© 2022. The Author(s).)

The detection and removal of poor-quality data in a training set is crucial to achieve high-performing AI models. In healthcare, data can be inherently poor-quality due to uncertainty or subjectivity, but as is often the case, the requirement for data privacy restricts AI practitioners from accessing raw training data, meaning manual visual verification of private patient data is not possible. Here we describe a novel method for automated identification of poor-quality data, called Untrainable Data Cleansing. This method is shown to have numerous benefits including protection of private patient data; improvement in AI generalizability; reduction in time, cost, and data needed for training; all while offering a truer reporting of AI performance itself. Additionally, results show that Untrainable Data Cleansing could be useful as a triage tool to identify difficult clinical cases that may warrant in-depth evaluation or additional testing to support a diagnosis., (© 2021. The Author(s).)

Histone deacetylase (HDAC) inhibitors are effective in MYCN-driven cancers, because of a unique need for HDAC recruitment by the MYCN oncogenic signal. However, HDAC inhibitors are much more effective in combination with other anti-cancer agents. To identify novel compounds which act synergistically with HDAC inhibitor, such as suberanoyl hydroxamic acid (SAHA), we performed a cell-based, high-throughput drug screen of 10,560 small molecule compounds from a drug-like diversity library and identified a small molecule compound (SE486-11) which synergistically enhanced the cytotoxic effects of SAHA. Effects of drug combinations on cell viability, proliferation, apoptosis and colony forming were assessed in a panel of neuroblastoma cell lines. Treatment with SAHA and SE486-11 increased MYCN ubiquitination and degradation, and markedly inhibited tumorigenesis in neuroblastoma xenografts, and, MYCN transgenic zebrafish and mice. The combination reduced ubiquitin-specific protease 5 (USP5) levels and increased unanchored polyubiquitin chains. Overexpression of USP5 rescued neuroblastoma cells from the cytopathic effects of the combination and reduced unanchored polyubiquitin, suggesting USP5 is a therapeutic target of the combination. SAHA and SE486-11 directly bound to USP5 and the drug combination exhibited a 100-fold higher binding to USP5 than individual drugs alone in microscale thermophoresis assays. MYCN bound to the USP5 promoter and induced USP5 gene expression suggesting that USP5 and MYCN expression created a forward positive feedback loop in neuroblastoma cells. Thus, USP5 acts as an oncogenic cofactor with MYCN in neuroblastoma and the novel combination of HDAC inhibitor with SE486-11 represents a novel therapeutic approach for the treatment of MYCN-driven neuroblastoma.

Study Question: Can an artificial intelligence (AI)-based model predict human embryo viability using images captured by optical light microscopy?, Summary Answer: We have combined computer vision image processing methods and deep learning techniques to create the non-invasive Life Whisperer AI model for robust prediction of embryo viability, as measured by clinical pregnancy outcome, using single static images of Day 5 blastocysts obtained from standard optical light microscope systems., What Is Known Already: Embryo selection following IVF is a critical factor in determining the success of ensuing pregnancy. Traditional morphokinetic grading by trained embryologists can be subjective and variable, and other complementary techniques, such as time-lapse imaging, require costly equipment and have not reliably demonstrated predictive ability for the endpoint of clinical pregnancy. AI methods are being investigated as a promising means for improving embryo selection and predicting implantation and pregnancy outcomes., Study Design, Size, Duration: These studies involved analysis of retrospectively collected data including standard optical light microscope images and clinical outcomes of 8886 embryos from 11 different IVF clinics, across three different countries, between 2011 and 2018., Participants/materials, Setting, Methods: The AI-based model was trained using static two-dimensional optical light microscope images with known clinical pregnancy outcome as measured by fetal heartbeat to provide a confidence score for prediction of pregnancy. Predictive accuracy was determined by evaluating sensitivity, specificity and overall weighted accuracy, and was visualized using histograms of the distributions of predictions. Comparison to embryologists' predictive accuracy was performed using a binary classification approach and a 5-band ranking comparison., Main Results and the Role of Chance: The Life Whisperer AI model showed a sensitivity of 70.1% for viable embryos while maintaining a specificity of 60.5% for non-viable embryos across three independent blind test sets from different clinics. The weighted overall accuracy in each blind test set was >63%, with a combined accuracy of 64.3% across both viable and non-viable embryos, demonstrating model robustness and generalizability beyond the result expected from chance. Distributions of predictions showed clear separation of correctly and incorrectly classified embryos. Binary comparison of viable/non-viable embryo classification demonstrated an improvement of 24.7% over embryologists' accuracy (P = 0.047, n = 2, Student's t test), and 5-band ranking comparison demonstrated an improvement of 42.0% over embryologists (P = 0.028, n = 2, Student's t test)., Limitations, Reasons for Caution: The AI model developed here is limited to analysis of Day 5 embryos; therefore, further evaluation or modification of the model is needed to incorporate information from different time points. The endpoint described is clinical pregnancy as measured by fetal heartbeat, and this does not indicate the probability of live birth. The current investigation was performed with retrospectively collected data, and hence it will be of importance to collect data prospectively to assess real-world use of the AI model., Wider Implications of the Findings: These studies demonstrated an improved predictive ability for evaluation of embryo viability when compared with embryologists' traditional morphokinetic grading methods. The superior accuracy of the Life Whisperer AI model could lead to improved pregnancy success rates in IVF when used in a clinical setting. It could also potentially assist in standardization of embryo selection methods across multiple clinical environments, while eliminating the need for complex time-lapse imaging equipment. Finally, the cloud-based software application used to apply the Life Whisperer AI model in clinical practice makes it broadly applicable and globally scalable to IVF clinics worldwide., Study Funding/competing Interest(s): Life Whisperer Diagnostics, Pty Ltd is a wholly owned subsidiary of the parent company, Presagen Pty Ltd. Funding for the study was provided by Presagen with grant funding received from the South Australian Government: Research, Commercialisation and Startup Fund (RCSF). 'In kind' support and embryology expertise to guide algorithm development were provided by Ovation Fertility. J.M.M.H., D.P. and M.P. are co-owners of Life Whisperer and Presagen. Presagen has filed a provisional patent for the technology described in this manuscript (52985P pending). A.P.M. owns stock in Life Whisperer, and S.M.D., A.J., T.N. and A.P.M. are employees of Life Whisperer., (© The Author(s) 2020. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology.)

Hypoxia-inducible factor (HIF)-1α accumulation promotes hematopoietic stem cells' quiescence and is necessary to maintain their self-renewal. However, the role of HIF-2α in hematopoietic cells is less clear. We investigated the role of HIF-2α in leukemia and lymphoma cells. HIF-2α expression was high in subsets of human and mouse leukemia and lymphoma cells, whereas it was low in normal bone marrow leukocytes. To investigate the role of HIF-2α, we transduced human HIF-2α cDNA in mouse syngeneic models of myeloid preleukemia and a transgenic model of B lymphoma. Ectopic expression of HIF-2α accelerated leukemia cell proliferation in vitro. Mice transplanted with cells transduced with HIF-2α died significantly faster of leukemia or B lymphoma than control mice transplanted with empty vector-transduced cells. Conversely, HIF-2α knockdown in human myeloid leukemia HL60 cells decreased proliferation in vitro and significantly prolonged animal survival following transplantation. In human acute myeloid leukemia (AML), HIF-2α mRNA was significantly elevated in several subsets such as the t(15;17), inv(16), complex karyotype and favorable cytogenetic groups. However, patients with high HIF-2α expression had a trend to higher disease-free survival in univariate analysis. The different effects of HIF-2α overexpression in mouse models of leukemia and human AML illustrates the complexity of this mutliclonal disease.

The majority of patients diagnosed with neuroblastoma present with aggressive disease. Improved detection of neuroblastoma cancer cells following initial therapy may help in stratifying patient outcome and monitoring for relapse. To identify potential plasma biomarkers, we utilised a liquid chromatography-tandem mass spectrometry-based proteomics approach to detect differentially-expressed proteins in serum from TH-MYCN mice. TH-MYCN mice carry multiple copies of the human MYCN oncogene in the germline and homozygous mice for the transgene develop neuroblastoma in a manner resembling the human disease. The abundance of plasma proteins was measured over the course of disease initiation and progression. A list of 86 candidate plasma biomarkers was generated. Pathway analysis identified significant association of these proteins with genes involved in the complement system. One candidate, complement C3 protein, was significantly enriched in the plasma of TH-MYCN(+/+) mice at both 4 and 6weeks of age, and was found to be elevated in a cohort of human neuroblastoma plasma samples, compared to healthy subjects. In conclusion, we have demonstrated the suitability of the TH-MYCN(+/+) mouse model of neuroblastoma for identification of novel disease biomarkers in humans, and have identified Complement C3 as a candidate plasma biomarker for measuring disease state in neuroblastoma patients., Biological Significance: This study has utilised a unique murine model which develops neuroblastoma tumours that are biologically indistinguishable from human neuroblastoma. This animal model has effectively allowed the identification of plasma proteins which may serve as potential biomarkers of neuroblastoma. Furthermore, the label-free ion count quantitation technique which was used displays significant benefits as it is less labour intensive, feasible and accurate. We have been able to successfully validate this approach by confirming the differential abundance of two different plasma proteins. In addition, we have been able to confirm that the candidate biomarker Complement C3, is more abundant in the plasma of human neuroblastoma patient plasma samples when compared to healthy counterparts. Overall we have demonstrated that this approach can be potentially useful in the identification of biomarker candidates, and that further validation of the candidates may lead to the discovery of novel, clinically useful diagnostic tools in the detection of sub-clinical neuroblastoma., (© 2013. Published by Elsevier B.V. All rights reserved.)

The mechanisms by which cells control their growth and behavioral identities are complex and require adaptability to environmental changes. Transcription factors act as master controllers of many of these pivotal points through their ability to influence the expression of many thousands of downstream genes, and increasingly research is showing that transcription factor regulation of target genes can change in response to environmental stimuli and cell type such that their function is not prescribed but rather context-dependent. Krüppel like factor 5 (KLF5) is an example of such a transcription factor, where evidence of disparate effects on cell growth and differentiation in normal and transformed tissue are clear. Here we present and discuss the literature covering the differential roles of KLF5 in particular tissues and cancer states, and the mechanisms by which these differences are effected through the regulation of KLF5 protein function in response to different cellular states and the direct effect on target gene expression., (© 2013 IUBMB.)

Krüppel-like factor 5 (KLF5) has been implicated as a tumor suppressor in various solid tumors such as breast and prostate, and recent studies have demonstrated a role for this protein in neutrophil differentiation of acute promyelocytic leukemia cells in response to ATRA. Here, we show that KLF5 expression increases during primary granulocyte differentiation and that expression of KLF5 is a requirement for granulocyte differentiation of 32D cells. In AML, we show that KLF5 mRNA expression levels are reduced in multiple French-American-British subtypes compared to normal controls, and also in leukemic stem cells relative to normal hematopoietic stem cells. We demonstrate that in selected AML cases, reduced expression is associated with hypermethylation of the KLF5 locus in the proximal promoter and/or intron 1, suggesting that this may represent a Class II genetic lesion in the development of AML., (Copyright © 2011 Elsevier Ltd. All rights reserved.)

The tumor suppressor Gadd45alpha was earlier shown to be a repressed target of sustained receptor-mediated ERK1/2 signaling. We have identified Gadd45alpha as a downregulated gene in response to constitutive signaling from two FLT3 mutants (FLT3-ITD and FLT3-TKD) commonly found in AML, and a leukemogenic GM-CSF receptor trans-membrane mutant (GMR-V449E). GADD45A mRNA downregulation is also associated with FLT3-ITD(+) AML. Sustained ERK1/2 signaling contributes significantly to receptor-mediated downregulation of Gadd45alpha mRNA in FDB1 cells expressing activated receptor mutants, and in the FLT3-ITD(+) cell line MV4;11. Knockdown of Gadd45alpha with shRNA led to increased growth and survival of FDB1 cells and enforced expression of Gadd45alpha in FDB1 cells expressing FLT3-ITD or GMR-V449E resulted in reduced growth and viability. Gadd45alpha overexpression in FLT3-ITD(+) AML cell lines also resulted in reduced growth associated with increased apoptosis and G(1)/S cell cycle arrest. Overexpression of Gadd45alpha in FDB1 cells expressing GMR-V449E was sufficient to induce changes associated with myeloid differentiation suggesting Gadd45alpha downregulation contributes to the maintenance of receptor-induced myeloid differentiation block. Thus, we show that ERK1/2-mediated downregulation of Gadd45alpha by sustained receptor signaling contributes to growth, survival and arrested differentiation in AML.

Mechanisms controlling the balance between proliferation and self-renewal versus growth suppression and differentiation during normal and leukemic myelopoiesis are not understood. We have used the bi-potent FDB1 myeloid cell line model, which is responsive to myelopoietic cytokines and activated mutants of the granulocyte macrophage-colony stimulating factor (GM-CSF) receptor, having differential signaling and leukemogenic activity. This model is suited to large-scale gene-profiling, and we have used a factorial time-course design to generate a substantial and powerful data set. Linear modeling was used to identify gene-expression changes associated with continued proliferation, differentiation, or leukemic receptor signaling. We focused on the changing transcription factor profile, defined a set of novel genes with potential to regulate myeloid growth and differentiation, and demonstrated that the FDB1 cell line model is responsive to forced expression of oncogenes identified in this study. We also identified gene-expression changes associated specifically with the leukemic GM-CSF receptor mutant, V449E. Signaling from this receptor mutant down-regulates CCAAT/enhancer-binding protein alpha (C/EBPalpha) target genes and generates changes characteristic of a specific acute myeloid leukemia signature, defined previously by gene-expression profiling and associated with C/EBPalpha mutations.