Revealing speckle obscured living human retinal cells with artificial intelligence assisted adaptive optics optical coherence tomography.

BACKGROUND: In vivo imaging of the human retina using adaptive optics optical coherence tomography (AO-OCT) has transformed medical imaging by enabling visualization of 3D retinal structures at cellular-scale resolution, including the retinal pigment epithelial (RPE) cells, which are essential for maintaining visual function. However, because noise inherent to the imaging process (e.g., speckle) makes it difficult to visualize RPE cells from a single volume acquisition, a large number of 3D volumes are typically averaged to improve contrast, substantially increasing the acquisition duration and reducing the overall imaging throughput. METHODS: Here, we introduce parallel discriminator generative adversarial network (P-GAN), an artificial intelligence (AI) method designed to recover speckle-obscured cellular features from a single AO-OCT volume, circumventing the need for acquiring a large number of volumes for averaging. The combination of two parallel discriminators in P-GAN provides additional feedback to the generator to more faithfully recover both local and global cellular structures. Imaging data from 8 eyes of 7 participants were used in this study. RESULTS: We show that P-GAN not only improves RPE cell contrast by 3.5-fold, but also improves the end-to-end time required to visualize RPE cells by 99-fold, thereby enabling large-scale imaging of cells in the living human eye. RPE cell spacing measured across a large set of AI recovered images from 3 participants were in agreement with expected normative ranges. CONCLUSIONS: The results demonstrate the potential of AI assisted imaging in overcoming a key limitation of RPE imaging and making it more accessible in a routine clinical setting.

The retinal pigment epithelium (RPE) is a single layer of cells within the eye that is crucial for vision. These cells are unhealthy in many eye diseases, and this can result in vision problems, including blindness. Imaging RPE cells in living human eyes is time consuming and difficult with the current technology. Our method substantially speeds up the process of RPE imaging by incorporating artificial intelligence. This enables larger areas of the eye to be imaged more efficiently. Our method could potentially be used in the future during routine eye tests. This could lead to earlier detection and treatment of eye diseases, and the prevention of some causes of blindness.

Social Determinants of Health in Community-Dwelling Dementia Patients Aged 65 and Over: Analysis of the 2019 National Health Interview Survey.

Alzheimer's Disease and related dementias affect 3.4 million community-dwelling adults in the United States. Given the burden of disease, a greater understanding of modifiable risk factors is crucial for targeted public health strategies. Social determinants of health (SDOH) are modifiable risk factors categorized in five domains: economic status, education, healthcare access, environment, and community context. Although individual SDOH have been linked to dementia, limited research exists on the interaction of SDOH with dementia across multiple domains. The aim of this study was to evaluate the association between SDOH across all five domains and dementia among community-dwelling adults in the United States. A cross-sectional study was performed on community-dwelling adults aged ≥65 years from the 2019 National Health Interview Survey (NHIS). Respondents (N = 9,277), of whom 303 (4%) self-reported positive dementia diagnosis, were predominantly female (55%), white (76%), and non-Hispanic (91%). Residing in a non-metropolitan area, having a usual place for healthcare, and receiving annual eye or dental exams were negatively associated with dementia. Minority compared to white status was not significantly associated with dementia, suggesting underdiagnosis of dementia within minority groups in the NHIS. We present the first comprehensive national view of SDOH among community-dwelling dementia patients in the United States.

Visualization of erythrocyte stasis in the living human eye in health and disease.

Blood cells trapped in stasis have been reported within the microcirculation, but their relevance to health and disease has not been established. In this study, we introduce an in vivo imaging approach that reveals the presence of a previously-unknown pool of erythrocytes in stasis, located within capillary segments of the CNS, and present in 100% of subjects imaged. These results provide a key insight that blood cells pause as they travel through the choroidal microvasculature, a vascular structure that boasts the highest blood flow of any tissue in the body. Demonstration of clinical utility using deep learning reveals that erythrocyte stasis is altered in glaucoma, indicating the possibility of more widespread changes in choroidal microvascular than previously realized. The ability to monitor the choroidal microvasculature at the single cell level may lead to novel strategies for tracking microvascular health in glaucoma, age-related macular degeneration, and other neurodegenerative diseases.

Widespread subclinical cellular changes revealed across a neural-epithelial-vascular complex in choroideremia using adaptive optics.

Choroideremia is an X-linked, blinding retinal degeneration with progressive loss of photoreceptors, retinal pigment epithelial (RPE) cells, and choriocapillaris. To study the extent to which these layers are disrupted in affected males and female carriers, we performed multimodal adaptive optics imaging to better visualize the in vivo pathogenesis of choroideremia in the living human eye. We demonstrate the presence of subclinical, widespread enlarged RPE cells present in all subjects imaged. In the fovea, the last area to be affected in choroideremia, we found greater disruption to the RPE than to either the photoreceptor or choriocapillaris layers. The unexpected finding of patches of photoreceptors that were fluorescently-labeled, but structurally and functionally normal, suggests that the RPE blood barrier function may be altered in choroideremia. Finally, we introduce a strategy for detecting enlarged cells using conventional ophthalmic imaging instrumentation. These findings establish that there is subclinical polymegathism of RPE cells in choroideremia.

Photoreceptor and Retinal Pigment Epithelium Relationships in Eyes With Vitelliform Macular Dystrophy Revealed by Multimodal Adaptive Optics Imaging.

Purpose: To assess the structure of cone photoreceptors and retinal pigment epithelial (RPE) cells in vitelliform macular dystrophy (VMD) arising from various genetic etiologies. Methods: Multimodal adaptive optics (AO) imaging was performed in 11 patients with VMD using a custom-assembled instrument. Non-confocal split detection and AO-enhanced indocyanine green were used to visualize the cone photoreceptor and RPE mosaics, respectively. Cone and RPE densities were measured and compared across BEST1-, PRPH2-, IMPG1-, and IMPG2-related VMD. Results: Within macular lesions associated with VMD, both cone and RPE densities were reduced below normal, to 37% of normal cone density (eccentricity 0.2 mm) and to 8.4% of normal RPE density (eccentricity 0.5 mm). Outside of lesions, cone and RPE densities were slightly reduced (both to 92% of normal values), but with high degree of variability in the individual measurements. Comparison of juxtalesional cone and RPE measurements (<1 mm from the lesion edge) revealed significant differences in RPE density across the four genes (P < 0.05). Overall, cones were affected to a greater extent than RPE in patients with IMPG1 and IMPG2 pathogenic variants, but RPE was affected more than cones in BEST1 and PRPH2 VMD. This trend was observed even in contralateral eyes from a subset of five patients who presented with macular lesions in only one eye. Conclusions: Assessment of cones and RPE in retinal locations outside of the macular lesions reveals a pattern of cone and RPE disruption that appears to be gene dependent in VMD. These findings provide insight into the cellular pathogenesis of disease in VMD.

Zandelisib with continuous or intermittent dosing as monotherapy or in combination with rituximab in patients with relapsed or refractory B-cell malignancy: a multicentre, first-in-patient, dose-escalation and dose-expansion, phase 1b trial.

BACKGROUND: Phosphatidylinositol 3-kinase p110δ (PI3Kδ) inhibitors are efficacious in B-cell malignancies. Immune-related adverse events might be mitigated with intermittent dosing. We aimed to evaluate the safety and antitumour activity of zandelisib, a potent novel PI3Kδ inhibitor, with continuous or intermittent dosing as monotherapy or in combination with rituximab, in patients with relapsed or refractory B-cell malignancy. METHODS: We conducted a first-in-patient, dose-escalation and dose-expansion, phase 1b trial at ten treatment centres across Switzerland and the USA. Eligible patients were aged 18 years or older with relapsed or refractory B-cell malignancy (limited to follicular lymphoma or chronic lymphocytic leukaemia during dose escalation) and an Eastern Cooperative Oncology Group performance status of 0-2, and had received at least one previous line of therapy and no previous PI3Kδ inhibitor treatment. In the dose-escalation study, participants received oral zandelisib once daily (60 mg, 120 mg, or 180 mg; we did not evaluate four additional planned dose levels). The 60 mg dose was further evaluated as monotherapy or with intravenous rituximab 375 mg/m2 on days 1, 8, 15, and 22 of cycle 1 and day 1 of cycles 3-6, using a continuous daily schedule or intermittent dosing therapy (days 1-28 of cycles 1-2 and days 1-7 of subsequent cycles) in 28-day cycles. Treatment was continued until evidence of disease progression, intolerance, or withdrawal of consent by the patient. Primary endpoints were safety (dose-limiting toxicities and maximum tolerated dose), minimum biologically effective dose, and a composite endpoint to assess the activity of each dose level, and were analysed by intention to treat. The zandelisib monotherapy and zandelisib-rituximab combination cohorts have completed accrual, but accrual to a cohort evaluating zandelisib with zanubrutinib is ongoing. This study is registered with ClinicalTrials.gov, NCT02914938. FINDINGS: Between Nov 17, 2016, and June 2, 2020, 100 patients were assessed for eligibility and 97 were enrolled and received zandelisib monotherapy (n=56) or zandelisib plus rituximab (n=41), with zandelisib administered on either a continuous schedule (n=38) or with intermittent dosing (n=59). No dose-limiting toxicities were observed, the objective of determining the maximum tolerated dose was abandoned, and antitumour activity was similar across the evaluated doses activity (objective responses in 11 [92%; 95% CI 61·5-99·8] of 12 patients at both 60 mg and 120 mg doses, and in five [83%; 95% CI 35·9-99·6] of six patients at 180 mg). With a median duration of exposure of 15·2 months (IQR 3·7-21·7), the most common grade 3-4 adverse events were neutrophil count decrease (ten [17%] of 59 patients in the intermittent dosing group and four [11%] of 38 in the continuous dosing group), diarrhoea (three [5%] and eight [21%]), pneumonia (one [2%] and six [16%]), alanine aminotransferase increase (three [5%] and two [5%]), and colitis (two [3%] and one [3%]). 26 (44%) of 59 patients in the intermittent dosing group and 29 (76%) of 38 patients in the continuous dosing group had grade 3-4 adverse events. Treatment-related serious adverse events occurred in eight (21%) patients in the continuous dosing group and five (8%) patients in the intermittent dosing group. There were no treatment-related deaths. INTERPRETATION: Zandelisib 60 mg once daily on an intermittent dosing schedule was safe, with low frequency of grade 3 or worse adverse events, warranting the ongoing global phase 2 and phase 3 trials. FUNDING: MEI Pharma.

Dietary interventions on the prevention and management of diabetes in post-kidney transplantation - A systematic review.

Post-transplant diabetes mellitus is associated with long-term immunosuppression and weight gain, and is related to an increased risk of cardiovascular disease, accelerated loss of graft and increased mortality. There is an absence of strong evidence-based dietary guidelines for the prevention and management of post-transplant diabetes mellitus in kidney transplant recipients. The aim of this study was to systematically review all dietary evidence for kidney transplant recipients on clinical outcomes relating to diabetes, patient-reported outcomes and economic outcomes. A comprehensive literature search was conducted in August 2020 using the databases Medline, Embase, CENTRAL and CINAHL. Studies were critically appraised using Cochrane risk of bias tools and GRADE. A total of 12 studies and 1928 participants were included. Four papers focused on diet and exercise, one paper on diet only, two papers on magnesium supplementation, one paper on magnesium and fibre intake, two papers on Mediterranean diet, one paper on marine n-3 fatty acid supplementation and one paper on fruit and vegetable intake. There were no significant effects on outcomes relating to dietary counselling, magnesium supplementation, magnesium and fibre intake or marine n-3 fatty acid supplementation. Low-quality evidence supports the Mediterranean diet in reducing the risk of post-transplant diabetes mellitus and fasting plasma glucose levels. Low-quality evidence suggests vegetable intake being associated with a lower risk of post-transplant diabetes mellitus. This review demonstrates limited evidence for dietary interventions in the prevention and management of diabetes in post-kidney transplantation. The findings suggest that further high-quality research with robust study designs is required.

Toward an index of oral somatosensory acuity: Comparison of three measures in adults.

PURPOSE: Somatosensory targets and feedback are instrumental in ensuring accurate speech production. Individuals differ in their ability to access and respond to somatosensory information, but there is no established standard for measuring somatosensory acuity. The primary objective of this study was to determine which of three measures of somatosensory acuity had the strongest association with change in production accuracy in a vowel learning task, while controlling for the better-studied covariate of auditory acuity. METHOD: Three somatosensory tasks were administered to 20 female college students: an oral stereognosis task, a bite block task with auditory masking, and a novel phonetic awareness task. Individual scores from the tasks were compared to their performance on a speech learning task in which participants were trained to produce novel Mandarin vowels with visual biofeedback. RESULTS: Of the three tasks, only bite block adaptation with auditory masking was significantly associated with performance in the speech learning task. Participants with weaker somatosensory acuity tended to demonstrate larger increases in production accuracy over the course of training. CONCLUSIONS: The bite block adaptation task measures proprioceptive awareness rather than tactile acuity and assesses somatosensory knowledge implicitly, with limited metalinguistic demands. This small-scale study provides preliminary evidence that these characteristics may be desirable for the assessment of oral somatosensory acuity, at least in the context of vowel learning tasks. Well-normed somatosensory measures could be of clinical utility by informing diagnosis/prognosis and treatment planning.

In-vivo sub-diffraction adaptive optics imaging of photoreceptors in the human eye with annular pupil illumination and sub-Airy detection.

Adaptive optics scanning light ophthalmoscopy (AOSLO) allows non-invasive visualization of the living human eye at the microscopic scale; but even with correction of the ocular wavefront aberrations over a large pupil, the smallest cells in the photoreceptor mosaic cannot always be resolved. Here, we synergistically combine annular pupil illumination with sub-Airy disk confocal detection to demonstrate a 33% improvement in transverse resolution (from 2.36 to 1.58 µm) and a 13% axial resolution enhancement (from 37 to 32 µm), an important step towards the study of the complete photoreceptor mosaic in heath and disease. Interestingly, annular pupil illumination also enhanced the visualization of the photoreceptor mosaic in non-confocal detection schemes such as split detection AOSLO, providing a strategy for enhanced multimodal imaging of the cone and rod photoreceptor mosaic.

Integrating adaptive optics-SLO and OCT for multimodal visualization of the human retinal pigment epithelial mosaic.

In vivo imaging of human retinal pigment epithelial (RPE) cells has been demonstrated through multiple adaptive optics (AO)-based modalities. However, whether consistent and complete information regarding the cellular structure of the RPE mosaic is obtained across these modalities remains uncertain due to limited comparisons performed in the same eye. Here, an imaging platform combining multimodal AO-scanning light ophthalmoscopy (AO-SLO) with AO-optical coherence tomography (AO-OCT) is developed to make a side-by-side comparison of the same RPE cells imaged across four modalities: AO-darkfield, AO-enhanced indocyanine green (AO-ICG), AO-infrared autofluorescence (AO-IRAF), and AO-OCT. Co-registered images were acquired in five subjects, including one patient with choroideremia. Multimodal imaging provided multiple perspectives of the RPE mosaic that were used to explore variations in RPE cell contrast in a subject-, location-, and even cell-dependent manner. Estimated cell-to-cell spacing and density were found to be consistent both across modalities and with normative data. Multimodal images from a patient with choroideremia illustrate the benefit of using multiple modalities to infer the cellular structure of the RPE mosaic in an affected eye, in which disruptions to the RPE mosaic may locally alter the signal strength, visibility of individual RPE cells, or even source of contrast in unpredictable ways.

Persistent Dark Cones in Oligocone Trichromacy Revealed by Multimodal Adaptive Optics Ophthalmoscopy.

Dark cone photoreceptors, defined as those with diminished or absent reflectivity when observed with adaptive optics (AO) ophthalmoscopy, are increasingly reported in retinal disorders. However, their structural and functional impact remain unclear. Here, we report a 3-year longitudinal study on a patient with oligocone trichromacy (OT) who presented with persistent, widespread dark cones within and near the macula. Diminished electroretinogram (ERG) cone but normal ERG rod responses together with normal color vision confirmed the OT diagnosis. In addition, the patient had normal to near normal visual acuity and retinal sensitivity. Occasional dark gaps in the photoreceptor layer were observed on optical coherence tomography, in agreement with reflectance AO scanning light ophthalmoscopy, which revealed that over 50% of the cones in the fovea were dark, increasing to 74% at 10° eccentricity. In addition, the cone density was 78% lower than normal histologic value at the fovea, and 20-40% lower at eccentricities of 5-15°. Interestingly, color vision testing was near normal at locations where cones were predominantly dark. These findings illustrate how a retina with predominant dark cones that persist over at least 3 years can support near normal central retinal function. Furthermore, this study adds to the growing evidence that cones can continue to survive under non-ideal conditions.

High-speed label-free two-photon fluorescence microscopy of metabolic transients during neuronal activity.

The brain is an especially active metabolic system, requiring a large supply of energy following neuronal activation. However, direct observation of cellular metabolic dynamics associated with neuronal activation is challenging with currently available imaging tools. In this study, an optical imaging approach combining imaging of calcium transients and the metabolic co-enzyme nicotinamide adenine dinucleotide (phosphate) (NAD(P)H) is utilized to track the metabolic dynamics in hippocampal neuron cultures. Results show distinct cellular components for the NAD(P)H response following neuronal activity, where notable differences in the NAD(P)H dynamics between neurons and astrocytes can be directly observed. Additionally, tracking of these responses across a large field of view is demonstrated for metabolic profiling of neuronal activation. Observation of neuronal dynamics using these methods allows for closer examination of the complex metabolic machinery of the brain, and may lead to a better understanding of the cellular metabolism of neuronal activation.

In vivo characterization of minipig skin as a model for dermatological research using multiphoton microscopy.

Minipig skin is one of the most widely used non-rodent animal skin models for dermatological research. A thorough characterization of minipig skin is essential for gaining deeper understanding of its structural and functional similarities with human skin. In this study, three-dimensional (3-D) in vivo images of minipig skin was obtained non-invasively using a multimodal optical imaging system capable of acquiring two-photon excited fluorescence (TPEF) and fluorescence lifetime imaging microscopy (FLIM) images simultaneously. The images of the structural features of different layers of the minipig skin were qualitatively and quantitatively compared with those of human skin. Label-free imaging of skin was possible due to the endogenous fluorescence and optical properties of various components in the skin such as keratin, nicotinamide adenine dinucleotide phosphate (NAD(P)H), melanin, elastin, and collagen. This study demonstrates the capability of optical biopsy techniques, such as TPEF and FLIM, for in vivo non-invasive characterization of cellular and functional features of minipig skin, and the optical image-based similarities of this commonly utilized model of human skin. These optical imaging techniques have the potential to become promising tools in dermatological research for developing a better understanding of animal skin models, and for aiding in translational pre-clinical to clinical studies.

Graded Image Generation Using Stratified CycleGAN.

In medical imaging, CycleGAN has been used for various image generation tasks, including image synthesis, image denoising, and data augmentation. However, when pushing the technical limits of medical imaging, there can be a substantial variation in image quality. Here, we demonstrate that images generated by CycleGAN can be improved through explicit grading of image quality, which we call stratified CycleGAN. In this image generation task, CycleGAN is used to upgrade the image quality and content of near-infrared fluorescent (NIRF) retinal images. After manual assignment of grading scores to a small subset of the data, semi-supervised learning is applied to propagate grades across the remainder of the data and set up the training data. These scores are embedded into the CycleGAN by adding the grading score as a conditional input to the generator and by integrating an image quality classifier into the discriminator. We validate the efficacy of the proposed stratified CycleGAN by considering pairs of NIRF images at the same retinal regions (imaged with and without correction of optical aberrations achieved using adaptive optics), with the goal being to restore image quality in aberrated images such that cellular-level detail can be obtained. Overall, stratified CycleGAN generated higher quality synthetic images than traditional CycleGAN. Evaluation of cell detection accuracy confirmed that synthetic images were faithful to ground truth images of the same cells. Across this challenging dataset, F1-score improved from 76.9 ± 5.7% when using traditional CycleGAN to 85.0±3.4% when using stratified CycleGAN. These findings demonstrate the potential of stratified Cycle-GAN to improve the synthesis of medical images that exhibit a graded variation in image quality.

Video-rate multimodal multiphoton imaging and three-dimensional characterization of cellular dynamics in wounded skin.

To date, numerous studies have been performed to elucidate the complex cellular dynamics in skin diseases, but few have attempted to characterize these cellular events under conditions similar to the native environment. To address this challenge, a three-dimensional (3D) multimodal analysis platform was developed for characterizing in vivo cellular dynamics in skin, which was then utilized to process in vivo wound healing data to demonstrate its applicability. Special attention is focused on in vivo biological parameters that are difficult to study with ex vivo analysis, including 3D cell tracking and techniques to connect biological information obtained from different imaging modalities. These results here open new possibilities for evaluating 3D cellular dynamics in vivo, and can potentially provide new tools for characterizing the skin microenvironment and pathologies in the future.

Tracking metabolic dynamics of apoptosis with high-speed two-photon fluorescence lifetime imaging microscopy.

Programmed cell death, or apoptosis, is an essential process in development and homeostasis, and disruptions in associated pathways are responsible for a wide variety of diseases such as cancer, developmental abnormalities, and Alzheimer's disease. On the other hand, cell death, in many cases, is the desired outcome of therapeutic treatments targeting diseases such as cancer. Recently, metabolic imaging based on two-photon fluorescence microscopy has been developed and shown to be highly sensitive to certain cell death processes, most notably apoptosis, thus having the potential as an advanced label-free screening tool. However, the typically low acquisition rates of this imaging technique have resulted in a limited throughput approach, allowing only a small population of cells to be tracked at well-separated time points. To address this limitation, a high-speed two-photon fluorescence lifetime imaging microscopy (2P-FLIM) platform capable of video-rate imaging is applied to study and further characterize the metabolic dynamics associated with cell death. Building upon previous work demonstrating the capabilities of this system, this microscope is utilized to study rapid metabolic changes during cell death induction, such as dose-dependency of metabolic response, response in invasive vs. noninvasive cancer cells, and response in an apoptosis-resistant cell line, which is further shown to undergo autophagy in response to toxic stimuli. Results from these experiments show that the early apoptosis-related metabolic dynamics are strongly correlated with important cellular parameters including responsiveness to apoptosis-inducing stimuli. The high speed and sensitivity of the presented imaging approach enables new investigations into this highly dynamic and complex process.

Cross-modal association between vowels and colours: A cross-linguistic perspective.

Previous studies showed similar mappings between sounds and colours for synaesthetes and non-synaesthetes alike, and proposed that common mechanisms underlie such cross-modal association. The findings between vowels and colours, and between pitch and lightness, were investigated separately, and it was also unknown how language background would influence such association. The present study investigated the cross-modal association between sounds (vowels and pitch) and colours in a tone language using three groups of non-synaesthetes: Cantonese (native), Mandarin (foreign, tonal), and English (foreign, non-tonal). Strong associations were found between /a/ and red, /i/ with light colours, and /u/ with dark colours, and a robust pitch effect with a high tone eliciting lighter colours than a low tone in general. The pitch effect is stronger than the vowel associations. Significant differences among the three language groups in colour choices of other vowels and the strength of association were found, which demonstrate the language-specificity of these associations. The findings support the notion that synaesthesia is a general phenomenon, which can be influenced by linguistic factors.

Intra-operative imaging of surgical margins of canine soft tissue sarcoma using optical coherence tomography.

Optical coherence tomography (OCT) is a rapid non-invasive imaging technique that has shown high sensitivity for intra-operative surgical margin assessment in human breast cancer clinical trials. This promising technology has not been evaluated in veterinary medicine. The objective of this study was to correlate normal and abnormal histological features with OCT images for surgical margins from excised canine soft tissue sarcoma (STS) and to establish image evaluation criteria for identifying positive surgical margins. Fourteen client-owned dogs underwent surgical resection of a STS and OCT imaging of 2 to 4 areas of interest on the resected specimen were performed. Following imaging these areas were marked with surgical ink and trimmed for histopathology evaluation. Results showed that different tissue types had distinct characteristic appearances on OCT imaging. Adipose tissue exhibited a relatively low scattering and a honey-comb texture pattern. Skeletal muscle and sarcoma tissue were both dense and highly scattering. While sarcoma tissue was highly scattering, it did not have organized recognizable structure in contrast to muscle which showed clear fibre alignment patterns. In this investigation, we showed different tissue types had different and characteristic scattering and image texture appearances on OCT, which closely correlate with low-power histology images. Given the differentiation between tissue types the results support that OCT could be used to identify positive surgical margins immediately following resection of STS. Further research is needed to assess the diagnostic accuracy of this method for surgical margin assessment.

Individual predictors of response to biofeedback training for second-language production.

While recent research suggests that visual biofeedback can facilitate speech production training in clinical populations and second language (L2) learners, individual learners' responsiveness to biofeedback is highly variable. This study investigated the hypothesis that the type of biofeedback provided, visual-acoustic versus ultrasound, could interact with individuals' acuity in auditory and somatosensory domains. Specifically, it was hypothesized that learners with lower acuity in a sensory domain would show greater learning in response to biofeedback targeting that domain. Production variability and phonological awareness were also investigated as predictors. Sixty female native speakers of English received 30 min of training, randomly assigned to feature visual-acoustic or ultrasound biofeedback, for each of two Mandarin vowels. On average, participants showed a moderate magnitude of improvement (decrease in Euclidean distance from a native-speaker target) across both vowels and biofeedback conditions. The hypothesis of an interaction between sensory acuity and biofeedback type was not supported, but phonological awareness and production variability were predictive of learning gains, consistent with previous research. Specifically, high phonological awareness and low production variability post-training were associated with better outcomes, although these effects were mediated by vowel target. This line of research could have implications for personalized learning in both L2 pedagogy and clinical practice.