P–263 Life Whisperer™, an AI-based algorithm to select non invasively best quality blastocysts for transfer: A multicenter analysis

2021 ◽  
Vol 36 (Supplement_1) ◽  
Author(s):  
P Muño. Espert ◽  
Y Galiana ◽  
L Medrano ◽  
J Ballester ◽  
L Ortega ◽  
...  
Keyword(s):  
Roc Curve ◽  
Inner Cell Mass ◽  
Cell Mass ◽  
High Sensitivity ◽  
Roc Curves ◽  
Time Lapse ◽  
Clinical Pregnancy ◽  
Embryo Selection ◽  
Computer Assisted ◽  
Stage Duration

Abstract Study question Is the AI-based Life Whisperer™ (LW) tool, suitable to evaluate blastocysts quality and predict clinical pregnancy (CP) in couples undergoing ICSI cycles? Summary answer LW blastocyst score is comparable to the scores of other classification methods. This AI model showed high sensitivity and a comparable specificity for CP. What is known already The morphology grading is the most widely used method for the selection and classification of the embryos in clinical practice.However,this evaluation entails intervariability and intravariability decision among the embryologists.Recently, research has been focused on new embryo selection systems based on computer-assisted evaluation such as time-lapse with complex algorithms that allow the recognition of objective parameters of the embryo morphology.The implementation of these technologies requires substantial investments that are not available for all clinics.LW is a new embryo selection method based on AI,where specific hardware is not needed,as it is based on single blastocyst images taken with a routine microscope. Study design, size, duration Between 2017–2020, a total of 513 Day–5 blastocysts, after ICSI, comming from egg donation treatment were included in this retrospective-multicentre study.Day–5 embryos were evaluated with 3 classification methods:Gardner’s blastocyst grade (GB), the computer derived-output Eeva (EV) and LW AI-supported system. The good quality blastocysts were first evaluated using the GB and EV scores and subsequently compared with the LW scores.The sensitivity and specificity of LW was assessed to validate this system as a clinical pregnancy predictor. Participants/materials, setting, methods A total of 513 Day–5 blastocysts, from 134 oocyte donation cycles, were evaluated first by GB score: expansion (1–6), inner cell mass and throphoectoderm (A-C).EV analyses the cell division timing P2 (2cells stage duration) and P3 (3cells stage duration) differentiating three categories:High,Medium and Low(VerMilyea et al.,2014).LW scores ranked 1–10 from a single Day–5 blastocyst HR Image performed on inverted microscope,with a threshold >5 for defining a viable blastocyst.T-test and ROC-curves were used for statistical analysis. Main results and the role of chance The average of LW score obtained from GB higher blastocyst expansion score (≥4) was 7.48±0.09, while the average of LW score obtained from GB lower blastocyst expansion score (<4) was 4.69±0.3 (P < 0.001). The average of LW score yielded from GB good morphology of Inner Cell Mass and trophoectoderm (AA,AB,BA) was 7.98±0.1 while the average of LW score obtained from GB lower quality blastocyst score (BB,BC,CB,CA,AC) was 6.36±0.156 (P < 0.001).The average of LW score resulted from EV High blastocysts was 7.42±0.17, while the average of this obtained from EV low score was 6.43±0.3 (P = 0.009).A correlation between EV and LW score could be assesed, except for the blastocyst that are considered Medium score from EV. Therefore, a strong correlation between GB and LW system, as well GB+EV and LW, was found and an equivalent usability of the LW tool could be confirmed. The analyse of LW score for transferred embryos (N = 156), using ROC curve, showed a high sensitivity (0,928) but a low specificity (0,154) with a threshold of 5. Regarding our data, ROC curve shows that a threshold of 8,46 could enhance the prediction of CPR because in this point the specifity value is higher than 0.5. Limitations, reasons for caution The LW score validation compared to GB and EV methodology was carried out on a small number of embryos.Additionally,not all embryos had been transferred at the time of the analysis.Thus to enhance the accuracy of these data and the specificity of the clinical prediction, a higher sample size is needed. Wider implications of the findings: 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. Additionally,the LW-system shows almost the highest sensibility and may also improve the specificity by self-learning feeding the AI-system, thus tailoring predictions to each laboratory unique environment. Trial registration number NA

scholarly journals Cytoplasmic String Between ICM and mTE Is a Positive Predictor of Clinical Pregnancy and Live Birth Outcomes in Elective Frozen-thawed Single Blastocyst Transfer Cycles: a Time-lapse Study

2020 ◽  
Author(s):  
Bing-Xin Ma ◽  
Lei Jin ◽  
Bo Huang
Keyword(s):  
Birth Outcomes ◽  
Live Birth ◽  
Inner Cell Mass ◽  
Cell Mass ◽  
Time Lapse ◽  
Clinical Pregnancy ◽  
Blastocyst Transfer ◽  
Single Blastocyst Transfer ◽  
First Time ◽  
Present Group

Abstract Background: In this study, we aim to investigate whether cytoplasmic string between inner cell mass (ICM) and mural trophectoderm (mTE) is a positive predictor of clinical pregnancy and live birth outcomes.Methods: 1,267 elective frozen-thawed single blastocyst transfer (eSBT) cycles cultured in time-lapse incubation system from January 2018 to May 2019 were involved in the study. Blastocysts were grouped according to the appearance of cytoplasmic strings between ICM and mTE cells, and identified as “Present” and “Absent” groups. In Present group, they were further categorized according to the quantity of cytoplasmic strings between ICM and mTE cells. Clinical pregnancy and live birth outcomes of blastocysts were used to evaluate the effect of cytoplasmic strings between ICM and mTE.Results: The baseline demographic and laboratory features were similar between the Present and Absent groups of cytoplasmic strings between ICM and mTE (P>0.05). According to the time-lapse analysis, cytoplasmic strings between ICM and mTE were more visible among good quality blastocysts. Furthermore, blastocysts with cytoplasmic strings showed a higher clinical pregnancy and live birth rates (P<0.05), and no significant differences were observed in abortion rate and birth weight (P>0.05).Conclusions: Although the previous conclusions of cytoplasmic strings were controversial, the present time-lapse analysis provides the evidence for the first time that cytoplasmic strings between ICM and mTE cells would be a positive predictor of clinical pregnancy and live birth outcomes in elective frozen-thawed single blastocyst transfer cycles.

scholarly journals Cytoplasmic String Between ICM and mTE Is a Positive Predictor of Clinical Pregnancy and Live Birth Outcomes in Elective Frozen-thawed Single Blastocyst Transfer Cycles: A Time-lapse Study

2021 ◽  
Author(s):  
Bingxin Ma ◽  
Lei Jin ◽  
Bo Huang
Keyword(s):  
Birth Outcomes ◽  
Live Birth ◽  
Inner Cell Mass ◽  
Cell Mass ◽  
Time Lapse ◽  
Clinical Pregnancy ◽  
Blastocyst Transfer ◽  
Single Blastocyst Transfer ◽  
First Time ◽  
Present Group

Abstract Background: In this study, we aim to investigate whether cytoplasmic string between inner cell mass (ICM) and mural trophectoderm (mTE) is a positive predictor of clinical pregnancy and live birth outcomes.Methods: 1,267 elective frozen-thawed single blastocyst transfer (eSBT) cycles cultured in time-lapse incubation system from January 2018 to May 2019 were involved in the study. Blastocysts were grouped according to the appearance of cytoplasmic strings between ICM and mTE cells, and identified as “Present” and “Absent” groups. In Present group, they were further categorized according to the quantity of cytoplasmic strings between ICM and mTE cells. Clinical pregnancy and live birth outcomes of blastocysts were used to evaluate the effect of cytoplasmic strings between ICM and mTE.Results: The baseline demographic and laboratory features were similar between the Present and Absent groups of cytoplasmic strings between ICM and mTE (P>0.05). According to the time-lapse analysis, cytoplasmic strings between ICM and mTE were more visible among good quality blastocysts. Furthermore, blastocysts with cytoplasmic strings showed a higher clinical pregnancy and live birth rates (P<0.05), and no significant differences were observed in abortion rate and birth weight (P>0.05).Conclusions: Although the previous conclusions of cytoplasmic strings were controversial, the present time-lapse analysis provides the evidence for the first time that cytoplasmic strings between ICM and mTE cells would be a positive predictor of clinical pregnancy and live birth outcomes in elective frozen-thawed single blastocyst transfer cycles.

P–132 Centralized versus local embryologists scoring of blastocyst quality obtained in a large European multicenter clinical trial

2021 ◽  
Vol 36 (Supplement_1) ◽  
Author(s):  
M Montag ◽  
E Va. de. Abbeel ◽  
T Ebner ◽  
P Larsson ◽  
B Mannaerts
Keyword(s):  
Quality Assessment ◽  
Data Storage ◽  
Inner Cell Mass ◽  
Cell Mass ◽  
Time Lapse ◽  
Ongoing Pregnancy Rate ◽  
Quality Classification ◽  
Blastocyst Quality ◽  
Selection Of ◽  
Local Assessment

Abstract Study question Does blastocyst quality scoring by central assessment deviate from local assessment and potentially lead to the selection of a different single blastocyst for transfer? Summary answer Central and local assessment provided the same quality classification (poor / good / top) in 69% of all blastocysts and 63% of all transferred blastocysts. What is known already Blastocyst quality is scored most frequently by three morphological parameters, namely expansion and hatching (EH) status, inner cell mass (ICM) grading and trophectoderm (TE) grading. The score is used to define the quality classification (poor / good / top) which determines which embryo is to be transferred or cryopreserved. Blastocyst scoring and grading can be highly subjective, which does influence the choice for transfer and cryopreservation. Time-lapse imaging technology captures additional input about embryo development as well as enables centralized data storage and sharing for independent central assessments. Study design, size, duration Pooled embryo analysis from a prospective, randomized, multicenter trial (RAINBOW) of 619 women undergoing ovarian stimulation with an individualized dose of follitropin delta in a long GnRH agonist protocol between May 2018 and January 2020. Blastocysts were centrally assessed using time-lapse images by two independent assessors and one adjudicator . Selection of the blastocyst for transfer by local assessment was based on morphological scoring and not on morphokinetic time-lapse parameters. Participants/materials, setting, methods Oocytes were fertilized by ICSI and cultured in the Embryoscopeâ (Vitrolife) up to day 5 for transfer or day 5/6 for cryopreservation. Embryos were assessed as either non-blastocyst or blastocyst. Blastocysts were graded centrally and locally at 116 hrs of development, based on EH status (1–6), ICM (A-D) and TE grading (A-D). Central assessors were blinded to local assessment and embryo transfer selection. Main results and the role of chance In total 4282 embryos were assessed centrally, of which 2046 day 5 embryos (48%) were adjudicated due to a scoring difference of at least one parameter between the two central assessors. In total 38% of day 5 embryos were judged as non-blastocysts and 62% as blastocysts of which 61% (i.e. 38% of all embryos) were determined to be of good or top quality. Identical results in terms of quality classification (poor / good / top) were obtained for 69% of blastocysts between local and central assessment and in 78%, between the two central assessors. Moreover, central and local scoring were identical in 62% for EH status, 53% for ICM grading and 57% for TE grading. For all transferred blastocysts (n = 508), central and local quality assessment was aligned for 63%. The ongoing pregnancy rate following single blastocyst transfer (SBT) was 41% (202/489), and similar to when considering only the transfers for which the central assessment had the same or a higher classification than the local assessment (166/411=40%). In 16% of all SBT, central quality assessment gave a lower score for the transferred blastocyst than the central assessment. This discrepancy could potentially have led to transfer of a different blastocyst. Limitations, reasons for caution This trial included assessments made by embryologists from 20 IVF centres. Some centres has limited experience with time-lapse technology for morphological blastocyst scoring. Scoring could therefore have been affected by differences in focal planes, magnification and contrast compared to inverted microscopy, with potential influence on blastocyst scores and quality classification. Wider implications of the findings: Local and central blastocyst quality classification based on morphology aligns well but remains subjective. Embryo assessment may benefit from using tools like artificial intelligence-based algorithms for a more objective analysis. Trial registration number NCT03564509

6 ANEUPLOIDY TOLERANCE IN RHESUS MACAQUE PRE-IMPLANTATION EMBRYOS VIA MICRONUCLEI FORMATION, CELLULAR FRAGMENTATION, AND BLASTOMERE EXCLUSION

2017 ◽  
Vol 29 (1) ◽  
pp. 110 ◽  
Author(s):  
B. L. Daughtry ◽  
J. L. Rosenkrantz ◽  
N. Lazar ◽  
N. Redmayne ◽  
K. A. Nevonen ◽  
...  
Keyword(s):  
Rhesus Macaque ◽  
Inner Cell Mass ◽  
Cell Mass ◽  
Time Lapse ◽  
Confocal Imaging ◽  
Chromosomal Mosaicism ◽  
Human Embryos ◽  
Cleavage Stage ◽  
Chromosomal Breakage

A primary contributor to in vitro fertilization (IVF) failure is the presence of unbalanced chromosomes in pre-implantation embryos. Previous array-based and next-generation sequencing (NGS) studies determined that ~50 to 80% of human embryos are aneuploid at the cleavage stage. During early mitotic divisions, many human embryos also sequester mis-segregated chromosomes into micronuclei and concurrently undergo cellular fragmentation. We hypothesised that cellular fragmentation represents a response to mis-segregated chromosomes that are encapsulated into micronuclei. Here, we utilised the rhesus macaque pre-implantation embryo as a model to study human embryonic aneuploidy using a combination of EevaTM time-lapse imaging for evaluating cell divisions, single-cell/-fragment DNA-Sequencing (DNA-Seq), and confocal microscopy of nuclear structures. Results from our time-lapse image analysis demonstrated that there are considerable differences in the timing of the first and third mitotic divisions between rhesus blastocysts and those that arrested before this stage in development (P < 0.01; ANOVA). By examining the chromosome content of each blastomere from cleavage stage embryos via DNA-Seq, we determined that rhesus embryos have an aneuploidy frequency up to ~62% (N = 26) with several embryos exhibiting chromosomal mosaicism between blastomeres (N = 6). Certain blastomeres also exhibited reciprocal whole chromosomal gains or losses, indicating that these embryos had undergone mitotic non-disjunction early in development. In addition, findings of reciprocal sub-chromosomal deletions/duplications among blastomeres suggest that chromosomal breakage had occurred in some embryos as well. Embryo immunostaining for the nuclear envelope protein, LAMIN-B1, demonstrated that fragmented cleavage-stage rhesus embryos often contain micronuclei and that cellular fragments can enclose DNA. Our DNA-Seq analysis confirmed that cellular fragments might encapsulate whole and/or partial chromosomes lost from blastomeres. When embryos were immunostained with gamma-H2AX, a marker of chromatin fragility, we observed distinct foci solely in micronuclei and DNA-containing cellular fragments. This suggests that micronuclei may be ejected from blastomeres through the process of cellular fragmentation and, once sequestered, these mis-segregated chromosomes become highly unstable and undergo DNA degradation. Finally, we also observed that ~10% of embryos prevented cellular fragments or large blastomeres from incorporating into the inner cell mass or trophectoderm at the blastocyst stage (n = 5). Upon confocal imaging, multiple nuclei and intense gamma-H2AX foci were found in a large unincorporated blastomere in one of the blastocysts. Altogether, our findings demonstrate that the rhesus embryo responds to segregation errors by eliminating chromosome-containing micronuclei via cellular fragmentation and/or selecting against aneuploid blastomeres that fail to divide during pre-implantation development with significant implications for human IVF.

scholarly journals Epigenetic Modification Affecting Expression of Cell Polarity and Cell Fate Genes to Regulate Lineage Specification in the Early Mouse Embryo

2010 ◽  
Vol 21 (15) ◽  
pp. 2649-2660 ◽  
Author(s):  
David-Emlyn Parfitt ◽  
Magdalena Zernicka-Goetz
Keyword(s):  
Cell Fate ◽  
Cell Polarity ◽  
Mouse Embryo ◽  
Inner Cell Mass ◽  
Epigenetic Modification ◽  
Cell Mass ◽  
Time Lapse ◽  
Cell Polarization ◽  
Asymmetric Divisions

Formation of inner and outer cells of the mouse embryo distinguishes pluripotent inner cell mass (ICM) from differentiating trophectoderm (TE). Carm1, which methylates histone H3R17 and R26, directs cells to ICM rather that TE. To understand the mechanism by which this epigenetic modification directs cell fate, we generated embryos with in vivo–labeled cells of different Carm1 levels, using time-lapse imaging to reveal dynamics of their behavior, and related this to cell polarization. This shows that Carm1 affects cell fate by promoting asymmetric divisions, that direct one daughter cell inside, and cell engulfment, where neighboring cells with lower Carm1 levels compete for outside positions. This is associated with changes to the expression pattern and spatial distribution of cell polarity proteins: Cells with higher Carm1 levels show reduced expression and apical localization of Par3 and a dramatic increase in expression of PKCII, antagonist of the apical protein aPKC. Expression and basolateral localization of the mouse Par1 homologue, EMK1, increases concomitantly. Increased Carm1 also reduces Cdx2 expression, a transcription factor key for TE differentiation. These results demonstrate how the extent of a specific epigenetic modification could affect expression of cell polarity and fate-determining genes to ensure lineage allocation in the mouse embryo.

P–158 Assisted Hatching on D + 3 in order to facilitate trophectoderm biopsy in blastocyst for PGT-A is not advisable in all patients

2021 ◽  
Vol 36 (Supplement_1) ◽  
Author(s):  
P Belchin ◽  
Y Cabello ◽  
M Sanche. d. Burgos ◽  
J Guerrero ◽  
M D Riva ◽  
...  
Keyword(s):  
Inner Cell Mass ◽  
Cell Mass ◽  
Time Lapse ◽  
Fertilization Rate ◽  
Blastocyst Stage ◽  
Assisted Hatching ◽  
Culture Dish ◽  
Key Factor ◽  
Registration Number ◽  
Number Of Cycles

Abstract Study question Is it useful or beneficial to perform Assisted Hatching (AH) on D + 3 previously to biopsy for PGT-A on blastocyst stage on D + 5? Summary answer The routine use of AH on D + 3 to facilitate the embryo biopsy on D + 5 could negatively influence the development of the embryos to blastocyst stage. What is known already The blastocyst stage is the optimal stage for performing biopsies for PGT-A, which has been reported as a key factor determining the growing clinical application of this strategy worldwide. For trophectoderm (TE) biopsy, laser-assisted drilling is used to create a zona opening on D + 3 or D + 5 of development. The method of zona opening on D + 3 allows some of the TE cells to herniate during blastocyst formation and expansion, which facilitates the biopsy process. However, this method may result in herniation of inner cell mass cells instead of TE or maybe could affect the development of the embryo to blastocyst stage. Study design, size, duration A total of 100 PGT-A cycles were performed in 2019 and 2020. In 78 of them laser-assisted drilling was used to create a zona opening on D + 5 only in those embryos which arrived to blastocyst stage for TE biopsy (Group No-AH). In 22 cycles the same drilling was achieved on D + 3 in all embryos, independently of their quality (Group AH). The average of embryos per cycle in each group was 5 and 4.3 respectively. Participants/materials, setting, methods A total of 100 PGT-A cycles coming from 65 patients were studied. The average of the age of the patients was 40.83 (SD 3.45) in the group No-AH vs 42.18 (SD 3.42) in the Group AH (p = 0.108), so the age was not a determining factor for the development of the embryos. We analyzed by χ 2 test differences between groups on fertilization rates, number of embryos, development to blastocyst stage, euploidy and pregnancy rates. Main results and the role of chance The fertilization rate was 74.79% (No-AH group) and 68.53% (AH group) with no significative statistical differences (p = 0.12). In the No-AH group, the TE biopsy was performed on D + 5 in 63 cycles (81%). In the AH group, 41% of cycles didn’t reach the blastocyst stage, obtaining statistical differences between groups (p = 0.035). We found also significant differences in the number of cycles with biopsied blastocyst when we had 1 to 6 embryos/cycle on D + 3 between groups (p = 0.002), without obtaining any blastocyst to be diagnosed in 53% of the cycles in AH group vs 27% in No-AH group. When the number of embryos on D + 3 per cycle was &gt; 6, at least 1 embryo reached the blastocyst stage in both groups, although this number was higher in No-AH group. The rate of biopsied blastocysts was significantly higher in the No-AH group compared to the AH group (46.61 vs 34.69) with a p = 0.031. The rate of euploid embryos analyzed was 23.30% in the No-AH group compared to 29.41% in the AH group, although no significant differences were found (p = 0.44) between groups. In the No-AH group, a clinical pregnancy rate of 52.94% was obtained (n = 34) vs 50% in the AH group (n = 4) (p = 0.91). Limitations, reasons for caution We have recently started to perform AH on D + 3, so the number of cases is smaller than No-AH group. We use a time lapse incubator in all cases, so in the No-AH the culture dish is changed, disturbing the stable incubation environment, while in the other group it is not. Wider implications of the findings: The use of AH on D + 3 in order to facilitate the TE biopsy on D + 5 could affect negatively the development of the embryos to blastocyst stage. Its routine use should be avoided based on laboratory workload, mainly if the patient has less than 7 embryos at D + 3. Trial registration number Not applicable

scholarly journals The Influence of Day 2 Blastomere Symmetry on Blastocyst Grade and Ploidy Status

2019 ◽  
Vol 01 (02) ◽  
pp. 115-118
Author(s):  
Diana Chieh Xing Tain ◽  
Michelle Sheng Rong Lim ◽  
Bee Lian Ng ◽  
Elizabeth Hammond ◽  
Pak Seng Wong
Keyword(s):  
Inner Cell Mass ◽  
Formation Rate ◽  
Cell Mass ◽  
Time Lapse ◽  
Blastocyst Stage ◽  
Chi Square ◽  
Cell Stage ◽  
Cell Asymmetry ◽  
Time Lapse Imaging ◽  
Ploidy Status

Previous studies have suggested that aneuploidy rates are co-related with cell asymmetry at the cleavage stage. A retrospective study was carried out to determine the significance of blastomere symmetry at the 4-cell stage on blastocyst grade and ploidy status. 732 Day 5/6 blastocysts from 191 patients undergoing Pre-implantation Genetic Testing for Aneuploidy were analysed with time-lapse imaging (Embryoscope, Vitrolife) during 2017. Blastomere symmetry was measured at the first image of 4-cells on Day 2 by tabulating the mean diameter of 2 lines drawn perpendicularly on each blastomere. Symmetry was defined as the blastomere diameter difference of [Formula: see text] 25%. Trophectoderm (TE) biopsy was performed on Day 5/6 followed by chromosomal evaluation using Next Generation Sequencing (VeriSeq Protocol, Illumina). Blastocyst grade was classified as either “Good” (inner cell mass (ICM) and TE, AA respectively), “Fair/Good” (AB, BA), “Fair” (BB) and “Poor” (early blastocyst grade 2 or TE grading of C). The significance of blastomere symmetry on blastocyst grade and ploidy status was measured using chi-square tests. There was no significance difference in resulting blastocyst quality for symmetrical and asymmetrical embryos (Table 1: p [Formula: see text] 0.10). Furthermore, there was no significance difference in the euploid rate (42.5% vs. 45.3%) or mosaic rate (22.1% vs. 16.2%) between symmetrical and asymmetrical embryos (p [Formula: see text] 0.24). In conclusion, the presence of asymmetrical blastomeres at the 4-cell stage do not impact the good quality blastocyst formation rate and euploidy rate for embryos that progress into blastocysts. However, this study excludes embryos that do not develop to the blastocyst stage and those with erratic division patterns, direct cleavage and reverse cleavage on Day 2, both of which have potential to influence ploidy result. Asymmetrical 4-cell embryos have the potential for high quality euploid blastocyst progression and can be considered for day 2 embryo transfer in the absence of symmetrical 4-cell embryos.

scholarly journals Evaluating the Performances of Biomarkers over a Restricted Domain of High Sensitivity

Mathematics ◽  
2021 ◽  
Vol 9 (21) ◽  
pp. 2826
Author(s):  
Manuel Franco ◽  
Juana-María Vivo
Keyword(s):  
Roc Curve ◽  
Performance Metrics ◽  
High Sensitivity ◽  
Roc Curves ◽  
Upper And Lower Bounds ◽  
Area Index ◽  
Colon Cancers ◽  
Partial Area ◽  
Variance Estimates ◽  
Inappropriate Expression

The burgeoning advances in high-throughput technologies have posed a great challenge to the identification of novel biomarkers for diagnosing, by contemporary models and methods, through bioinformatics-driven analysis. Diagnostic performance metrics such as the partial area under the ROC (pAUC) indexes exhibit limitations to analysing genomic data. Among other issues, the inability to differentiate between biomarkers whose ROC curves cross each other with the same pAUC value, the inappropriate expression of non-concave ROC curves, and the lack of a convenient interpretation, restrict their use in practice. Here, we have proposed the fitted partial area index (FpAUC), which is computable through an algorithm valid for any ROC curve shape, as an alternative performance summary for the evaluation of highly sensitive biomarkers. The proposed approach is based on fitter upper and lower bounds of the pAUC in a high-sensitivity region. Through variance estimates, simulations, and case studies for diagnosing leukaemia, and ovarian and colon cancers, we have proven the usefulness of the proposed metric in terms of restoring the interpretation and improving diagnostic accuracy. It is robust and feasible even when the ROC curve shows hooks, and solves performance ties between competitive biomarkers.

Sign in / Sign up

Export Citation Format

Share Document