Scientific publications and presentations


Scientific publications and presentations

Published scientific studies and peer-reviewed conference proceedings. Also includes scientific presentations and summaries of clinical data.

Featured publications and conferences:

  • Human Reproduction

  • Nature Scientific Reports

  • Reproductive BioMedicine Online

  • ASPIRE: 2023

  • PCRS: 2023
  • ESHRE: 2019, 2020, 2021, 2022, 2023

  • ASRM: 2019, 2020, 2022, 2023

Jump to section

Life Whisperer Viability | Life Whisperer GeneticsLife Whisperer Oocytes | Life Whisperer Embryo Viability and Genetics | Machine Learning Technology

Published scientific studies and peer-reviewed conference proceedings. Also includes scientific presentations and summaries of clinical data.

Featured publications and conferences:

  • Human Reproduction

  • Nature Scientific Reports

  • Reproductive BioMedicine Online

  • ASPIRE: 2023

  • PCRS: 2023

  • ESHRE: 2019, 2020, 2021, 2022, 2023

  • ASRM: 2019, 2020, 2022, 2023

Jump to section

Life Whisperer Viability

Life Whisperer Genetics

Life Whisperer Oocytes

Life Whisperer Embryo Viability and Genetics

Machine Learning Technology

 

Life Whisperer Viability


An artificial intelligence model correlated with morphological and genetic features of blastocyst quality improves ranking of viable embryos (RBMO 2022)

Winner of the Reproductive Biomedicine Online annual Robert G. Edwards Prize Paper Award for the best paper published in 2022

Improved methods for evaluating artificial intelligence in the field of IVF are described. The importance of correlating intelligence scores with known parameters of embryo quality for artificial intelligence characterization is highlighted. The findings support artificial intelligence testing methods and use in clinical practice.


Development of an artificial intelligence-based assessment model for prediction of embryo viability using static images captured by optical light microscopy during IVF (Hum Reprod 2020)

We have combined computer vision image processing methods and deep learning techniques to create the noninvasive 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.


Improved time to pregnancy when combining an artificial intelligence score and morphology grade for embryo selection during IVF (ASPIRE 2023)

LWV showed improved embryo ranking and reduction in the estimated average number of transfers needed to achieve clinical pregnancy. Furthermore, evaluation of time-to-pregnancy supports the combined use of LWV+Gardner grading, showing that they work synergistically to further improve ranking performance when selecting average quality embryos that may be transferred in later cycles.


An artificial intelligence algorithm outperforms highly variable embryologist grading for predicting the likelihood of pregnancy outcome from embryo images (ASPIRE 2023)

This study demonstrates the inherent variability and lack of objectivity associated with an embryologist’s evaluation of embryos. It highlights the benefits of accurate AI algorithms for standardizing embryo assessment.


Comparative analysis of pregnancy predictive potential using the deep learning blastocyst scoring model calculated from the single focus blastocyst image and time-lapse image sequences (P-08 ASPIRE 2023)


This study identified that pregnancy rates increased according to higher scores for both Life Whisperer Viability and iDAscore. Life Whisperer, which calculates scores from a single focus and single time-point blastocyst image: only 1 image, has comparable ability to predict pregnant blastocysts compared to iDAscore, which uses time-lapse image sequences to calculate scores considering morphokinetics: total of 7920 images.


Prediction of implantation of vitrified-warmed blastocysts using a deep learning algorithm on a single post-warming image of each embryo (O-158 ESHRE 2023)


The use of predictive models on vitrified cycles may help to select the best frozen embryo for transfer and predict which embryos will result in implantation failure early enough to thaw another one with a better chance of success. It also indicates the AI can generalize to thawed embryo assessment.


An artificial intelligence model that was trained on pregnancy outcomes for embryo viability assessment is highly correlated with Gardner Score (O-222 ESHRE 2021)

AI provides additional information over Gardner about pregnancy outcomes.


AI-based assessment of embryo viability correlates with features of embryo ploidy (P-228 ESHRE 2021)

The AI score correlated with genetic features of embryos that are known to correlate with pregnancy, which further supports the efficacy and use of AI for embryo viability assessment. Combination of embryo viability and embryo genetic assessment is a powerful measure of embryo quality and is likely to improve embryo ranking and pregnancy outcomes.


Past embryo viability is not always a good predictor of future pregnancy: dynamic viability suggests video has limited benefit over static images for AI assessment (P-202 ESHRE 2021)

The findings suggest that static end-point AI assessment is sufficient for predicting embryo implantation potential. Continual AI monitoring of embryos can potentially allow optimisation of which embryo to transfer and when, to ultimately improve pregnancy outcomes for patients


Life Whisperer, an AI-based algorithm to non-invasively select best quality blastocysts for transfer: A multicenter analysis (P-263 ESHRE 2021)


Blastocyst selection looks equivalent between all systems, but the LW tool is more objective and faster, saving time and costs significantly, without needing substantial hardware investments. The LW system shows almost the highest sensitivity and may also improve the specificity by self-learning feeding the AI-system, thus tailoring predictions to each laboratory unique environment.


Evidence for superior blastocyst cohort ranking using artificial intelligence based on retrospective clinical pregnancy results (P-92 ASRM 2020)

An AI model trained on clinical pregnancy data showed superior ranking ability and a shorter TTP compared with embryologists’ ranking (random chance), for simulated cohorts of transferred embryos.


Camera-agnostic self-annotating Artificial Intelligence (AI) system for blastocyst evaluation (O-116 ESHRE 2020)

Automated annotation and deep learning were successfully applied to create scalable AI that work with both standard microscope images and time-lapse images.


Artificial intelligence (AI) technology can predict human embryo viability across multiple laboratories with varying demographics with high accuracy and reproducibility (O-004 ESHRE 2020)

This comprehensive analysis in multiple clinical environments, spanning various microscope and camera equipment, and in different demographic locations, provides strong evidence that Life Whisperer AI delivers an improved predictive accuracy in classifying embryo viability. The implication of these results is that the AI can be used to inform selection of embryos, suggesting that this improved accuracy in selection of the best embryo for a given patient will result in improved pregnancy success rates and reduce the overall number of cycles leading to a pregnancy for a given patient.

 

Life Whisperer Genetics


Development of an artificial intelligence model for predicting the likelihood of human embryo euploidy based on blastocyst images from multiple imaging systems during IVF (Hum Reprod 2022)

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.


Simulated cohort analyses showed that an Automatic AI score reduced the number of cycles needed to achieve live birth and increased the first-cycle success rates (O-97 ASRM 2023)

The analysis of one 120h-embryo image by Artificial Intelligence may be indicative of its potential to lead to pregnancy and a live birth, and to reduce the miscarriage rates.


Applying Intelligence Artificial to embryos: can a deep learning algorithm predict the chromosomal status of an embryo? (O-96 ASRM 2023)

The non-invasive AI algorithm was able to predict embryo euploidy with an AUC of 0.75. The study showed the capability of this new algorithm based on AI in distinguishing between euploid and aneuploid embryos with apparently normal morphology.


An artificial intelligence algorithm demonstrates optimal performance for evaluating embryo genetic status at 120 hours post-fertilization (P-144 ESHRE 2023)

These results suggest that the AI is providing additional information regarding embryo genetic status, over and above that provided by known morphological parameters. The dynamic nature of AI score related to expansion is of interest as it relates to the optimal time-point for conducting analyses for selection of euploid embryos.


Prediction of live birth using an artificial intelligence (AI) algorithm developed to evaluate embryo genetic status from day 5 images (P-8 ASRM 2022)

A genetics AI for evaluating embryo images was predictive of live birth, suggesting it may be detecting morphological features correlated with genetic status that are also indicative of live birth outcome.


Artificial intelligence: Non-invasive detection of morphological features associated with abnormalities in chromosomes 21 and 16 (P-329 ASRM 2019)

For the first time, we have shown that AI can non-invasively detect morphological features of human Day 5 blastocysts with specific chromosomal abnormalities.

 

Life Whisperer Oocytes


A non-invasive artificial intelligence (AI) algorithm can predict competence of denuded oocytes from images taken prior to intracytoplasmic sperm injection (ICSI) (P-4 ASRM 2022)

A novel AI algorithm had high predictive power for assessing whether oocytes will develop into a usable blastocyst from single, static oocyte images, denuded prior to ICSI. Further, a UDC data cleansing technique was able to improve AI performance by identifying and removing cases where good quality oocytes were likely mislabeled as noncompetent due to external factors beyond oocyte quality, such as male infertility.

 

Life Whisperer Embryo Viability and Genetics


Development of a non-invasive artificial intelligence algorithm for identification of euploid embryos with high morphological quality during IVF (O-179 ASRM 2023)

The EQ score is highly predictive for identifying euploid embryos with high morphological quality. The combined score was comparable to individual genetics and viability AI scores for predicting PGT-A and pregnancy outcomes, respectively. 


Development of an automated system for morphological classification of embryos using image-based artificial intelligence (O-176 ASRM 2023)

An automated, AI system for embryo classification is likely to significantly improve standardization and efficiency in the IVF laboratory, providing more consistent grading results, and reducing the time spent on manual grading. This system could be used to support regional reporting requirements.


Development of a combined artificial intelligence score for evaluating both embryo ploidy and viability to aid in embryo selection during IVF (ASPIRE 2023)

An AI score that can evaluate both embryo ploidy and viability simultaneously is useful for selecting preferred embryos for analysis or transfer. These results suggest that it is feasible to generate a single score for evaluating overall embryo quality using a non-invasive approach.


Combined use of artificial intelligence (AI) algorithms for evaluating embryo viability and embryo genetics improves selection of embryos leading to clinical pregnancy (P-5 ASRM 2022)

Pre-selection of embryos using genetics AI improved subsequent ranking using viability AI, with fewer cycles needed to achieve pregnancy. Results suggest genetics AI may be used in a similar manner to PGT-A to preselect embryos that are more likely to be euploid, followed by morphology-based selection for transfer.


AI study shows the effect of patient age on embryo quality is inherent in the morphology of an embryo (P-234 ESHRE 2022)

Age-related effects on embryo quality are inherently captured in embryo morphology. AI algorithms that assess morphology correlate with expected decline in embryo quality with age.


Embryo viability and non-invasive genetic assessment through the lens of AI (The Best of ASRM and ESHRE 2021)

Selecting the best embryo in IVF is critical to a successful pregnancy outcome.

 

Machine Learning Technology


Efficient automated error detection in medical data using deep‑learning and label‑clustering (Nature Sci Rep 2023)

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.


A novel decentralized federated learning approach to train on globally distributed, poor quality, and protected private medical data (Nature Sci Rep 2022)

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.


Automated detection of poor‑quality data: case studies in healthcare (Nature Sci Rep 2021)

 Here we describe a novel method for automated identification of poor-quality data, called Untrainable Data Cleansing (UDC). 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.


The future of the AI-enabled digital lab must be driven by global collaboration (The Best of ASRM and ESHRE 2021)

Connected globally diverse data can produce the world’s most powerful AI that is unbiased, scalable, accessible and affordable.


A natural language processing approach of global survey results on what the embryologist thinks and faces (P-782 ESHRE 2021)

Embryologists, essential professionals of Fertility Centres, are less satisfied in many quantifiable aspects, but they love their profession and have many aspirational goals.

Contact Us

For scientific enquiries please email info@lifewhisperer.com with the subject line 'Scientific Enquiry'.