Image-based Identification Research Articles

Automated volumetric analysis of the inner ear fluid space from hydrops magnetic resonance imaging using 3D neural networks

Due to the development of magnetic resonance (MR) imaging processing technology, image-based identification of endolymphatic hydrops (EH) has played an important role in understanding inner ear illnesses, such as Meniere’s disease or fluctuating sensorineural hearing loss. We segmented the inner ear, consisting of the cochlea, vestibule, and semicircular canals, using a 3D-based deep neural network model for accurate and automated EH volume ratio calculations. We built a dataset of MR cisternography (MRC) and HYDROPS-Mi2 stacks labeled with the segmentation of the perilymph fluid space and endolymph fluid space of the inner ear to devise a 3D segmentation deep neural network model. End-to-end learning was used to segment the perilymph fluid and the endolymph fluid spaces simultaneously using aligned pair data of the MRC and HYDROPS-Mi2 stacks. Consequently, the segmentation performance of the total fluid space and endolymph fluid space had Dice similarity coefficients of 0.9574 and 0.9186, respectively. In addition, the EH volume ratio calculated by experienced otologists and the EH volume ratio value predicted by the proposed deep learning model showed high agreement according to the interclass correlation coefficient (ICC) and Bland–Altman plot analysis.

YOLOv5s-Based Image Identification of Stripe Rust and Leaf Rust on Wheat at Different Growth Stages.

Stripe rust caused by Puccinia striiformis f. sp. tritici and leaf rust caused by Puccinia triticina, are two devastating diseases on wheat, which seriously affect the production safety of wheat. Timely detection and identification of the two diseases are essential for taking effective disease management measures to reduce wheat yield losses. To realize the accurate identification of wheat stripe rust and wheat leaf rust during the different growth stages, in this study, the image-based identification of wheat stripe rust and wheat leaf rust during different growth stages was investigated based on deep learning using image processing technology. Based on the YOLOv5s model, we built identification models of wheat stripe rust and wheat leaf rust during the seedling stage, stem elongation stage, booting stage, inflorescence emergence stage, anthesis stage, milk development stage, and all the growth stages. The models were tested on the different testing sets in the different individual growth stages and in all the growth stages. The results showed that the models performed differently in disease image identification. The model based on the disease images acquired during an individual growth stage was not suitable for the identification of the disease images acquired during the other individual growth stages, except for the model based on the disease images acquired during the milk development stage, which had acceptable identification performance on the testing sets in the anthesis stage and the milk development stage. In addition, the results demonstrated that wheat growth stages had a great influence on the image identification of the two diseases. The model built based on the disease images acquired in all the growth stages produced acceptable identification results. Mean F1 Score values between 64.06% and 79.98% and mean average precision (mAP) values between 66.55% and 82.80% were achieved on each testing set composed of the disease images acquired during an individual growth stage and on the testing set composed of the disease images acquired during all the growth stages. This study provides a basis for the image-based identification of wheat stripe rust and wheat leaf rust during the different growth stages, and it provides a reference for the accurate identification of other plant diseases.

An intelligent sensing platform for detecting and identifying biochemical substances based on terahertz spectra

This paper presents the development of an intelligent sensing platform dedicated to accurately identifying terahertz (THz) spectra obtained from various biochemical substances. The platform currently has two distinct identification modes, which focus on identifying five amino acids, namely phenylalanine, methionine, lysine, leucine, and threonine, and five carbohydrates, namely aspartame, fructose, glucose, lactose monohydrate, and sucrose based on their THz spectra. The first mode, called One-dimensional THz Spectrum Identification (OTSI), combines THz time-domain spectroscopy (THz-TDS) with the proposed mini convolutional neural network (MCNN) model. THz-TDS detects biochemical substances, while the MCNN model identifies the THz spectra. The MCNN model has a simple structure and only needs to deal with the THz absorption coefficients of biochemical substances, which are less computationally intensive and easily converged. The model can achieve 99.07 % accuracy in identifying one-dimensional THz spectra of the ten biochemical substances. The second mode, THz Spectrum Image-based Identification (TSII), applies the YOLO-v5 target detection model to THz spectral image recognition. The YOLO-v5 model uses THz absorption peaks as identification features and can identify biochemical substances based on only one or several THz absorption peaks. The overall identifying accuracy of the YOLO-v5 model for ten biochemical substances is 96.20 %. We also compared the MCNN and YOLO-v5 models with other deep learning and machine learning models, which demonstrate that they have better performance. This feature broadens the platform's utility in biomolecular analysis and paves the way for further research and development in detecting and analyzing diverse biological compounds.

Image-based identification and isolation of micronucleated cells to dissect cellular consequences.

Recent advances in isolating cells based on visual phenotypes have transformed our ability to identify the mechanisms and consequences of complex traits. Micronucleus (MN) formation is a frequent outcome of genome instability, triggers extensive disease-associated changes in genome structure and signaling coincident with MN rupture, and is almost exclusively defined by visual analysis. Automated MN detection in microscopy images has proved extremely challenging, limiting unbiased discovery of the mechanisms and consequences of MN formation and rupture. In this study we describe two new MN segmentation modules: a rapid and precise model for classifying micronucleated cells and their rupture status (VCS MN), and a robust model for accurate MN segmentation (MNFinder) from a broad range of microscopy images. As a proof-of-concept, we define the transcriptome of non-transformed human cells with intact or ruptured MN after inducing chromosome missegregation by combining VCS MN with photoactivation-based cell isolation and RNASeq. Surprisingly, we find that neither MN formation nor rupture triggers a unique transcriptional response. Instead, transcriptional changes are correlated with increased aneuploidy in these cell classes. Our MN segmentation modules overcome a significant challenge to reproducible MN quantification, and, joined with visual cell sorting, enable the application of powerful functional genomics assays, including pooled CRISPR screens and time-resolved analyses of cellular and genetic consequences, to a wide-range of questions in MN biology.

Thoracic computed tomography (CT) image-based identification and severity classification of COVID-19 cases using vision transformer (ViT)

In this research, we developed a two-stage deep learning (DL) model using Vision Transformer (ViT) to detect COVID-19 and assess its severity from thoracic CT images. In the first stage, we utilized a pre-trained ViT model (ViT_B/32) and a custom CNN model to classify CT images as COVID-19 or non-COVID-19. The ViT model achieved superior performance with a fivefold cross-validated accuracy of 99.7%, compared to the custom CNN’s 98%. In the second stage, we employed a ViT-based U-Net model (Vision Transformer for Biomedical Image Segmentation, VITBIS) to segment lung and infection regions in COVID-19 positive CT images, determining the infection severity. This model uses transformers with attention mechanisms in both the encoder and decoder. The lung segmentation network achieved an Intersection Over Union (IOU) of 95.8% and a sensitivity of 99.67%, while the lesion segmentation network attained an IOU of 94% and a sensitivity of 98.3%.

Image-based recognition of parasitoid wasps using advanced neural networks.

Hymenoptera has some of the highest diversity and number of individuals among insects. Many of these species potentially play key roles as food sources, pest controllers and pollinators. However, little is known about the diversity and biology and ~80% of the species have not yet been described. Classical taxonomy based on morphology is a rather slow process but DNA barcoding has already brought considerable progress in identification. Innovative methods such as image-based identification and automation can further speed up the process. We present a proof of concept for image data recognition of a parasitic wasp family, the Diapriidae (Hymenoptera), obtained as part of the GBOL III project. These tiny (1.2-4.5mm) wasps were photographed and identified using DNA barcoding to provide a solid ground truth for training a neural network. Taxonomic identification was used down to the genus level. Subsequently, three different neural network architectures were trained, evaluated and optimised. As a result, 11 different genera of diaprids and one mixed group of 'other Hymenoptera' can be classified with an average accuracy of 96%. Additionally, the sex of the specimen can be classified automatically with an accuracy of >97%.

Image-based identification of optical quality and functional properties in inkjet-printed electronics using machine learning

AbstractWe propose a novel image-analysis based machine-learning approach to the fully-automated identification of the optical quality, of functional properties, and of manufacturing parameters in the field of 2D inkjet-printed test structures of conductive traces. To this end, a customizable modular concept to simultaneously identify or predict dissimilar properties of printed functional structures based on images is described and examined. An application domain of the concept in the printing production process is outlined. To examine performance, we develop a dataset of over 5000 test structures containing images and physical characteristics, which are manufactured using commercially available materials. Functional test structures are fabricated via a single-nozzle vector-based inkjet-printing system and thermally sintered. Physical characterization of electrical conductance, image capturing, and evaluation of the optical quality of the test structures is done by an automatic in-house built measurement station. Conceptionally, the design of a convolutional neural network is described to identify the optical quality and physical characteristics based only on acquired images. A mathematical apparatus that allows assessment of the identification accuracy is developed and described. The impact of printing resolution, of emerging defects in the geometry of printed structures, and of image quality and color space on the identification accuracy is analyzed. Quality groups related to the printing resolution that affect identification accuracy are determined. Supplementarily, we introduce not yet reported classification of processes related to the fabrication of printed functional structures, adopted from the process analytical technology.

A review of the Attagenus (Coleoptera: Dermestidae) species from South Africa and an image-based key to identification

The Dermestidae is a relatively poorly studied family of insects but with a high rate of species discovery. The South African region is particularly rich in Attagenus spp. many of which are distinctively patterned. This study provides an updated list of Attagenus spp. in South Africa with a complete literature review and an image-based identification key to the Attagenus species within South Africa. Images of most species are provided. With the rise in significance of citizen science recording schemes, this study facilitates image-based identification and encourages both entomologists and citizen scientists to further contribute to the understanding of this beautiful Coleopteran family.

Comparing modern identification methods for wild bees: Metabarcoding and image-based morphological taxonomic assignment.

With the decline of bee populations worldwide, studies determining current wild bee distributions and diversity are increasingly important. Wild bee identification is often completed by experienced taxonomists or by genetic analysis. The current study was designed to compare two methods of identification including: (1) morphological identification by experienced taxonomists using images of field-collected wild bees and (2) genetic analysis of composite bee legs (multiple taxa) using metabarcoding. Bees were collected from conservation grasslands in eastern Iowa in summer 2019 and identified to the lowest taxonomic unit using both methods. Sanger sequencing of individual wild bee legs was used as a positive control for metabarcoding. Morphological identification of bees using images resulted in 36 unique taxa among 22 genera, and >80% of Bombus specimens were identified to species. Metabarcoding was limited to genus-level assignments among 18 genera but resolved some morphologically similar genera. Metabarcoding did not consistently detect all genera in the composite samples, including kleptoparasitic bees. Sanger sequencing showed similar presence or absence detection results as metabarcoding but provided species-level identifications for cryptic species (i.e., Lasioglossum). Genus-specific detections were more frequent with morphological identification than metabarcoding, but certain genera such as Ceratina and Halictus were identified equally well with metabarcoding and morphology. Genera with proportionately less tissue in a composite sample were less likely to be detected using metabarcoding. Image-based methods were limited by image quality and visible morphological features, while genetic methods were limited by databases, primers, and amplification at target loci. This study shows how an image-based identification method compares with genetic techniques, and how in combination, the methods provide valuable genus- and species-level information for wild bees while preserving tissue for other analyses. These methods could be improved and transferred to a field setting to advance our understanding of wild bee distributions and to expedite conservation research.

Machine Learning Approaches for the Image-Based Identification of Surgical Wound Infections: Scoping Review.

Surgical site infections (SSIs) occur frequently and impact patients and health care systems. Remote surveillance of surgical wounds is currently limited by the need for manual assessment by clinicians. Machine learning (ML)-based methods have recently been used to address various aspects of the postoperative wound healing process and may be used to improve the scalability and cost-effectiveness of remote surgical wound assessment. The objective of this review was to provide an overview of the ML methods that have been used to identify surgical wound infections from images. We conducted a scoping review of ML approaches for visual detection of SSIs following the JBI (Joanna Briggs Institute) methodology. Reports of participants in any postoperative context focusing on identification of surgical wound infections were included. Studies that did not address SSI identification, surgical wounds, or did not use image or video data were excluded. We searched MEDLINE, Embase, CINAHL, CENTRAL, Web of Science Core Collection, IEEE Xplore, Compendex, and arXiv for relevant studies in November 2022. The records retrieved were double screened for eligibility. A data extraction tool was used to chart the relevant data, which was described narratively and presented using tables. Employment of TRIPOD (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis) guidelines was evaluated and PROBAST (Prediction Model Risk of Bias Assessment Tool) was used to assess risk of bias (RoB). In total, 10 of the 715 unique records screened met the eligibility criteria. In these studies, the clinical contexts and surgical procedures were diverse. All papers developed diagnostic models, though none performed external validation. Both traditional ML and deep learning methods were used to identify SSIs from mostly color images, and the volume of images used ranged from under 50 to thousands. Further, 10 TRIPOD items were reported in at least 4 studies, though 15 items were reported in fewer than 4 studies. PROBAST assessment led to 9 studies being identified as having an overall high RoB, with 1 study having overall unclear RoB. Research on the image-based identification of surgical wound infections using ML remains novel, and there is a need for standardized reporting. Limitations related to variability in image capture, model building, and data sources should be addressed in the future.

A novel integrated strategy of easy pruning, parameter searching, and re-parameterization for lightweight intelligent lithology identification

Lithology identification is a necessary task for activities of reservoir evaluation and underground engineering construction. Recently, the intelligent lithology identification method based on deep learning of image has attracted more and more attention with its excellent performance. However, the existing models are not developed specifically for lithology identification, and there is still room for improving accuracy and a large number of redundant calculations during inference. Therefore, it is necessary to develop a lightweight model specifically for lithologic identification. In order to realize accurate, rapid and effective lithology identification, we propose a novel integrated strategy of easy pruning, parameter searching, and reparameterization for lightweight intelligent lithology identification. Firstly, a model dedicated to lithology classification and easy to pruning is constructed by using the multi-branch structure; secondly, a pruning parameter search method based on differential evolution of expressivity is proposed; and finally, the model is further compressed using reparameterization. We validate the entire method using 160 lithologies. P3DNet has an accuracy of 97.69% on 160 lithologies. The pruning method based on differential evolution of expressivity can realize the optimal pruning under the condition of predetermined accuracy. The P3DNet with 134,380 parameters can achieve an accuracy of 97.50% on 160 lithologies, and it takes only 3.84 ms for the NVIDIA GeForce GTX 2080 Ti GPU to infer an image with a resolution of 224×224×3. The pruned P3DNet is 39.54% faster than MobileNet v3 and 50.65% faster than ShuffleNet V2. Our method can significantly reduce the requirements of the image-based identification method on device memory, computing power, etc., and greatly expand the application scenarios of the intelligent identification method. Relevant research can provide strong support for rapid and intelligent identification of lithology in site survey and construction.

Individual identification of endangered amphibians using deep learning and smartphone images: case study of the Japanese giant salamander (Andrias japonicus)

Information obtained via individual identification is invaluable for ecology and conservation. Physical tags, such as PIT tags and GPS, have been used for individual identification; however, these methods could impact on animal behavior and survival rates, and the tags may become lost. Although non-invasive methods that do not affect the target species (such as manual photoidentification) are available, these techniques utilize stripes and spots that are unique to the individual, which requires training, and applying them to large datasets is challenging. Many studies that have applied deep learning for identification have focused on species-level identification, but few have addressed individual-level identification. In this study, we developed an image-based identification method based on deep learning that uses the head spot pattern of the Japanese giant salamander (Andrias japonicus), an endemic and endangered species in Japan. We trained and evaluated a dataset collected over two days from 11 individuals in captivity, which included 7075 images taken by a smartphone camera. Individuals were photographed three times a day at approximately 11:00 (morning), 15:00 (afternoon), and 18:00 (evening). As a result, individual identification by our method, which used the EfficientNetV2 achieved 99.86% accuracy, kappa coefficient of 0.99, and an F1 score of 0.99. Performance was lower for the evening model than for the morning and afternoon models, which were trained and evaluated using photographs taken at the corresponding time of the day. The proposed method does not require direct contact with the target species, and the effect on the animals is minimal; moreover, individual-level information can be obtained under natural conditions. In the future, smartphone images can be applied to citizen science surveys and individual-level big data collection, which is difficult using current methods.

Comparative Study: Evaluating the effects of class balancing on transformer performance in the PlantNet-300k image dataset

Image-based identification of plant specimens plays a crucial role in various fields such as agriculture, ecology, and biodiversity conservation. The growing interest in deep learning has led to remarkable advancements in image classification techniques, particularly with the utilization of convolutional neural networks (CNNs). Since 2015, in the context of the PlantCLEF (Conference and Labs of the Evaluation Forum) challenge (Joly et al. 2015), deep learning models, specifically CNNs, have consistently achieved the most impressive results in this field (Carranza-Rojas 2018). However, recent developments have introduced transformer-based models, such as ViT (Vision Transformer) (Dosovitskiy et al. 2020) and CvT (Convolutional vision Transformer) (Wu et al. 2021), as a promising alternative for image classification tasks. Transformers offer unique advantages such as capturing global context and handling long-range dependencies (Vaswani et al. 2017), which make them suitable for complex recognition tasks like plant identification. In this study, we focus on the image classification task using the PlantNet-300k dataset (Garcin et al. 2021a). The dataset consists of a large collection of 306,146 plant images representing 1,081 distinct species. These images were selected from the Pl@ntNet citizen observatory database. The dataset has two prominent characteristics that pose challenges for classification. First, there is a significant class imbalance, meaning that a small subset of species dominates the majority of the images. This imbalance creates bias and affects the accuracy of classification models. Second, many species exhibit visual similarities, making it tough, even for experts, to accurately identify them. These characteristics are referred to by the dataset authors as long-tailed distribution and high intrinsic ambiguity, respectively (Garcin et al. 2021b). In order to address the inherent challenges of the PlantNet-300k dataset, we employed a two-fold approach. Firstly, we leveraged transformer-based models to tackle the dataset's intrinsic ambiguity and effectively capture the complex visual patterns present in plant images. Secondly, we focused on mitigating the class imbalance issue through various data preprocessing techniques, specifically class balancing methods. By implementing these techniques, we aimed to ensure fair representation of all plant species in order to improve the overall performance of image classification models. Our objective is to assess the effects of data preprocessing techniques, specifically class balancing, on the classification performance of the PlantNet-300k dataset. By exploring different preprocessing methods, we addressed the class imbalance issue and through precise evaluation, conducted a comparison of the performance of transformer-based models with and without class balancing techniques. Through these efforts, our ultimate goal is to assert if these techniques allow us to achieve more accurate and reliable classification results, particularly for underrepresented species in the dataset. In our experiment, we compared the performance of two transformer-based models, ViT and CvT, using two versions of the PlantNet-300k dataset: one with class balancing and the other without class balancing. This setup results in a total of four sets of metrics for evaluation. To assess the classification performance, we utilized a wide range of commonly used metrics including recall, precision, accuracy, AUC (Area Under the Curve), ROC (Receiver Operating Characteristic), and others. These metrics provide insights into each models' ability to correctly classify plant species, identify false positives and negatives, measure overall accuracy, and assess the models' discriminatory power. By conducting this comparative study, we seek to contribute to the advancement of plant identification research by providing empirical evidence of the benefits and effectiveness of class balancing techniques in improving the performance of transformer-based models on the PlantNet-300k dataset and any other similar ones.

Machine learning analysis of wing venation patterns accurately identifies Sarcophagidae, Calliphoridae and Muscidae fly species.

In medical, veterinary and forensic entomology, the ease and affordability of image data acquisition have resulted in whole-image analysis becoming an invaluable approach for species identification. Krawtchouk moment invariants are a classical mathematical transformation that can extract local features from an image, thus allowing subtle species-specific biological variations to be accentuated for subsequent analyses. We extracted Krawtchouk moment invariant features from binarised wing images of 759 male fly specimens from the Calliphoridae, Sarcophagidae and Muscidae families (13 species and a species variant). Subsequently, we trained the Generalized, Unbiased, Interaction Detection and Estimation random forests classifier using linear discriminants derived from these features and inferred the species identity of specimens from the test samples. Fivefold cross-validation results show a 98.56 ± 0.38% (standard error) mean identification accuracy at the family level and a 91.04 ± 1.33% mean identification accuracy at the species level. The mean F1-score of 0.89 ± 0.02 reflects good balance of precision and recall properties of the model. The present study consolidates findings from previous small pilot studies of the usefulness of wing venation patterns for inferring species identities. Thus, the stage is set for the development of a mature data analytic ecosystem for routine computer image-based identification of fly species that are of medical, veterinary and forensic importance.

Drying kinetics and image-based identification of drying end point during parboiling of komal chawal

During the process of parboiling, steamed rice goes through a process of drying till a suitable moisture content for storage. This work is based on classifying suitable moisture content of less than 13% for storage. Further, this classification is used for estimating the end point of drying based on images of rice taken out at different time intervals. The drying process is a basic process for food technologists and involves mathematical modeling to determine the process parameters. Optimal drying time calculation to achieve desired moisture content involves the use of mathematical modeling. There are two ways of calculating the end point. One is using already existing drying models, e.g., Page, Newton, etc., with time as a factor; the other is using FF- ANN with one hidden layer with drying time and temperature as a factor; and for theoretical understating of the diffusion process, the analytical solution of Fick's law was studied. The other method involves using extrinsic data like the RGB values and pixel values of images mapped with moisture content classes predicted using machine learning models. This study aims to show how RGB data-based machine learning models for classification gives real time solutions to problems like calculating the final moisture content after drying of parboiled brown rice. Digital images were used as input, pixel values were used as features, and the images were labeled into two categories: A (moisture greater than 13%) and B (less than or equal to 13%). For this purpose, the dataset used consists of the moisture content of samples at different times (0–180 min) and temperatures (40, 50, and 60 °C) of drying in a tray dryer. Semi empirical models were evaluated, and Modified-Page showed the best fitting, The Feed forward Artificial Neural Network was observed to show an accuracy of 0.882 based on the least square error. A convolutional neural network (CNN) was used to map the image pixel data to classify the rice sample based on moisture content with a validation accuracy of 0.79 and a test accuracy of 0.74. Similarly, RGB values-based ML models like CART showed better accuracy of 0.92 approximately. Lastly, a simple calculator for an android application was built in MIT App Inventor by selecting the best fitted model of classification (LR, KNN, NB, or SVM) based on RGB values as the features.

An Image-Based Identification of Aggressive Breast Cancer Circulating Tumor Cell Subtypes.

Using previously established CTC lines from breast cancer patients, we identified different morphometric subgroups of CTCs with one of them having the highest tumorigenic potential in vivo despite the slowest cell proliferation in vitro. This subgroup represents 32% of all cells and contains cells with small cell volume, large nucleus to cell, dense nuclear areas to the nucleus, mitochondria to cell volume ratios and rough texture of cell membrane and termed "Small cell, Large mitochondria, Rough membrane" (SLR). RNA-seq analyses showed that the SLR group is enriched in pathways and cellular processes related to DNA replication, DNA repair and metabolism. SLR upregulated genes are associated with poor survival in patients with ER+ breast cancer based on the KM Plotter database. The high tumorigenic potential, slow proliferation, and enriched DNA replication/repair pathways suggest that the SLR subtype is associated with stemness properties. Our new findings provide a simple image-based identification of CTC subpopulations with elevated aggressiveness, which is expected to provide a more accurate prediction of patient survival and therapy response than total CTC numbers. The detection of morphometric and transcriptomic profiles related to the SLR subgroup of CTCs also opens opportunities for potential targeted cancer treatment.

Use of Postmortem Computed Tomography in Disaster Victim Identification: Current Japanese Methods and Challenges

Introduction:In Japan, victims of large-scale disasters are usually identified by non-objective means. In the case of the 2011 Great East Japan Earthquake, ~90% of the bodies were identified based on nonobjective means such as facial features or belongings, which resulted in misidentification. At present, the situation remains the same. However, according to global standards, a method referred to as “disaster victim identification” (DVI; individual identification of disaster victims) is recommended by the International Criminal Police Organization; in this method, a multidisciplinary investigation team integrates objective information such as dental charts and DNA. Furthermore, recently, there has been a movement to employ postmortem computed tomography (CT) for personal identification, and radiologists are expected to be included in the DVI team.Method:In the Department of Legal Medicine of Chiba University in Japan, individual identification via CT or magnetic resonance imaging was conducted in forensic autopsy cases of unknown identities when there was an assumed person for the body and the antemortem image of the person could be acquired. Two certified radiologists interpreted and compared the antemortem CT with the postmortem CT taken prior to autopsy and assessed whether the two images were compatible to indicate the same person.Results:A total of 20 cases were judged. In all cases, two images were compatible, indicating the same person. Image-based identification was particularly useful when dental findings or fingerprints were unavailable for comparison and there were no family members available for DNA testing.Conclusion:In the future, this method will be applied to large-scale disasters.

Abstract CT104: MC1R imaging and histology in the targeted imaging of melanoma for alpha-particle radiotherapy (TIMAR1) trial

Abstract Purpose: Melanocortin 1 receptor (MC1R) has been shown in animal models to be a target for effective radionuclide therapy. This first-in-human clinical trial studied the safety and biodistribution of two novel MC1R-targeted imaging tracers designed to support the development of MC1R-targeted alpha particle therapy through image-based identification of MC1R and radiation dosimetry. Imaging was compared to MC1R expression by immunohistochemistry (IHC) of subjects’ melanoma tumors. Methods: Subjects with stage IV melanoma were randomized to receive either [203Pb]VMT01 SPECT/CT imaging followed by [68Ga]VMT02 PET/CT imaging approximately 1 month later, or vice versa in a cross-over design during their normal melanoma treatment course. Patient selection included a positive [18F]FDG PET/CT within 30 days of imaging tracer injection. 555 - 925 mBq of [203Pb]VMT01 was injected IV with imaging performed at 1, 4, and 24 hours. 74 - 277 mBq [68Ga]VMT02 was injected IV and imaging performed dynamically up to 1 hour, and at 2 and 3 hours. Due to an interruption in drug supply, only 3 subjects received [203Pb]VMT01 imaging. Key FDG-avid melanoma lesions were marked and presented to three experienced Radiologists serving as blinded reviewers. The blinded reviewers compared [18F]FDG PET/CT to experimental imaging and scored scans and key lesions on multiple metrics. Imaging positivity was defined as tumor uptake and retention of tracer in ≥ 1 melanoma tumors above background liver activity. Melanoma biopsy tissue was obtained and MC1R IHC performed as available. Results: 7 subjects were enrolled; 6 were imaged and 5 had available pathology. 3 of 6 imaged subjects had MC1R positive tumors via experimental imaging tracers. [68Ga]VMT02 PET/CT at 3 hours led to the best tumor to background ratio, and [203Pb]VMT01 SPECT/CT showed tumor retention at 24 hours. 4 of 5 subjects had positive MC1R staining by IHC. Of the 3 subjects with negative imaging, 2 had positive staining by IHC. Of the 3 subjects with positive imaging, 2 had available biopsies and all stained positive for MC1R by IHC. Conclusion: A subgroup of advanced melanoma patients were identified to express MC1R using new MC1R-targeted imaging agents. Full concordance between imaging and IHC of positive scans, but not with negative scans, suggests that IHC is a more sensitive biomarker detection method. Additional testing is needed to establish frequency of positive MC1R imaging in stage IV melanoma and its correlation to histological findings and treatment outcomes. Preferential tumor retention provides proof-of-concept that the MC1R-targeting therapy trial for [212Pb]VMT01 alpha-emitting therapy will deliver localized radiation to high MC1R expressing tumors as screened by [203Pb]VMT01 or [68Ga]VMT02. Imaging quality is sufficient for subject selection and dosimetry but may exclude subjects with MC1R receptor expression only detectable by IHC. Citation Format: Brenna M. Marks, Matthew Block, Geoffrey B. Johnson, Carrie B. Hruska, Mukesh K. Pandey, Andrew Paulsen, Michael K. Schultz, Michael A. McDonald, Edwin A. Sagastume, Frances L. Johnson. MC1R imaging and histology in the targeted imaging of melanoma for alpha-particle radiotherapy (TIMAR1) trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr CT104.

Label-free liquid biopsy through the identification of tumor cells by machine learning-powered tomographic phase imaging flow cytometry

Image-based identification of circulating tumor cells in microfluidic cytometry condition is one of the most challenging perspectives in the Liquid Biopsy scenario. Here we show a machine learning-powered tomographic phase imaging flow cytometry system capable to provide high-throughput 3D phase-contrast tomograms of each single cell. In fact, we show that discrimination of tumor cells against white blood cells is potentially achievable with the aid of artificial intelligence in a label-free flow-cyto-tomography method. We propose a hierarchical machine learning decision-maker, working on a set of features calculated from the 3D tomograms of the cells’ refractive index. We prove that 3D morphological features are adequately distinctive to identify tumor cells versus the white blood cell background in the first stage and, moreover, in recognizing the tumor type at the second decision step. Proof-of-concept experiments are shown, in which two different tumor cell lines, namely neuroblastoma cancer cells and ovarian cancer cells, are used against monocytes. The reported results allow claiming the identification of tumor cells with a success rate higher than 97% and with an accuracy over 97% in discriminating between the two cancer cell types, thus opening in a near future the route to a new Liquid Biopsy tool for detecting and classifying circulating tumor cells in blood by stain-free method.

Identification of fish spermatogenic cells through high-throughput immunofluorescence against testis with an antibody set.

Image-based identification and quantification of different types of spermatogenic cells is of great importance, not only for reproductive studies but also for genetic breeding. Here, we have developed antibodies against spermatogenesis-related proteins in zebrafish (Danio rerio), including Ddx4, Piwil1, Sycp3, and Pcna, and a high-throughput method for immunofluorescence analysis of zebrafish testicular sections. By immunofluorescence analysis of zebrafish testes, our results demonstrate that the expression of Ddx4 decreases progressively during spermatogenesis, Piwil1 is strongly expressed in type A spermatogonia and moderately expressed in type B spermatogonia, and Sycp3 has distinct expression patterns in different subtypes of spermatocytes. Additionally, we observed polar expression of Sycp3 and Pcna in primary spermatocytes at the leptotene stage. By a triple staining of Ddx4, Sycp3, and Pcna, different types/subtypes of spermatogenic cells were easily characterized. We further demonstrated the practicality of our antibodies in other fish species, including Chinese rare minnow (Gobiocypris rarus), common carp (Cyprinus carpio), blunt snout bream (Megalobrama amblycephala), rice field eel (Monopterus albus) and grass carp (Ctenopharyngodon idella). Finally, we proposed an integrated criterion for identifying different types/subtypes of spermatogenic cells in zebrafish and other fishes using this high-throughput immunofluorescence approach based on these antibodies. Therefore, our study provides a simple, practical, and efficient tool for the study of spermatogenesis in fish species.