Lauren M. Zuromski, Jacob Durtschi, Aimal Aziz, Jeffrey Chumley, Mark Dewey, Paul English, Muir Morrison, Keith Simmon, Blaine Whipple, Brendan O'Fallon, David P. Ng
{"title":"Clinical Validation of a Real-Time Machine Learning-based System for the Detection of Acute Myeloid Leukemia by Flow Cytometry","authors":"Lauren M. Zuromski, Jacob Durtschi, Aimal Aziz, Jeffrey Chumley, Mark Dewey, Paul English, Muir Morrison, Keith Simmon, Blaine Whipple, Brendan O'Fallon, David P. Ng","doi":"arxiv-2409.11350","DOIUrl":"https://doi.org/arxiv-2409.11350","url":null,"abstract":"Machine-learning (ML) models in flow cytometry have the potential to reduce\u0000error rates, increase reproducibility, and boost the efficiency of clinical\u0000labs. While numerous ML models for flow cytometry data have been proposed, few\u0000studies have described the clinical deployment of such models. Realizing the\u0000potential gains of ML models in clinical labs requires not only an accurate\u0000model, but infrastructure for automated inference, error detection, analytics\u0000and monitoring, and structured data extraction. Here, we describe an ML model\u0000for detection of Acute Myeloid Leukemia (AML), along with the infrastructure\u0000supporting clinical implementation. Our infrastructure leverages the resilience\u0000and scalability of the cloud for model inference, a Kubernetes-based workflow\u0000system that provides model reproducibility and resource management, and a\u0000system for extracting structured diagnoses from full-text reports. We also\u0000describe our model monitoring and visualization platform, an essential element\u0000for ensuring continued model accuracy. Finally, we present a post-deployment\u0000analysis of impacts on turn-around time and compare production accuracy to the\u0000original validation statistics.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"49 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142266667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Dynamic landscapes and statistical limits on growth during cell fate specification","authors":"Gautam Reddy","doi":"arxiv-2409.09548","DOIUrl":"https://doi.org/arxiv-2409.09548","url":null,"abstract":"The complexity of gene regulatory networks in multicellular organisms makes\u0000interpretable low-dimensional models highly desirable. An attractive geometric\u0000picture, attributed to Waddington, visualizes the differentiation of a cell\u0000into diverse functional types as gradient flow on a dynamic potential\u0000landscape, but it is unclear under what biological constraints this metaphor is\u0000mathematically precise. Here, we show that growth-maximizing regulatory\u0000strategies that guide a single cell to a target distribution of cell types are\u0000described by time-dependent potential landscapes, under certain generic\u0000growth-control tradeoffs. Our analysis leads to a sharp bound on the time it\u0000takes for a population to grow to a target distribution of a certain size. We\u0000show how the framework can be used to compute Waddington-like epigenetic\u0000landscapes and growth curves in an illustrative model of growth and\u0000differentiation. The theory suggests a conceptual link between nonequilibrium\u0000thermodynamics and cellular decision-making during development.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"20 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142266668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"(Un)buckling mechanics of epithelial monolayers under compression","authors":"Chandraniva Guha Ray, Pierre A. Haas","doi":"arxiv-2409.07928","DOIUrl":"https://doi.org/arxiv-2409.07928","url":null,"abstract":"When cell sheets fold during development, their apical or basal surfaces\u0000constrict and cell shapes approach the geometric singularity in which these\u0000surfaces vanish. Here, we reveal the mechanical consequences of this geometric\u0000singularity for tissue folding in a minimal vertex model of an epithelial\u0000monolayer. In simulations of the buckling of the epithelium under compression\u0000and numerical solutions of the corresponding continuum model, we discover an\u0000\"unbuckling\" bifurcation: At large compression, the buckling amplitude can\u0000decrease with increasing compression. By asymptotic solution of the continuum\u0000equations, we reveal that this bifurcation comes with a large stiffening of the\u0000epithelium. Our results thus provide the mechanical basis for absorption of\u0000compressive stresses by tissue folds such as the cephalic furrow during\u0000germband extension in Drosophila.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"32 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yuri G. Vilela, Artur C. Fassoni, Armando G. M. Neves
{"title":"On the design and stability of cancer adaptive therapy cycles: deterministic and stochastic models","authors":"Yuri G. Vilela, Artur C. Fassoni, Armando G. M. Neves","doi":"arxiv-2409.06867","DOIUrl":"https://doi.org/arxiv-2409.06867","url":null,"abstract":"Adaptive therapy is a promising paradigm for treating cancers, that exploits\u0000competitive interactions between drug-sensitive and drug-resistant cells,\u0000thereby avoiding or delaying treatment failure due to evolution of drug\u0000resistance within the tumor. Previous studies have shown the mathematical\u0000possibility of building cyclic schemes of drug administration which restore\u0000tumor composition to its exact initial value in deterministic models. However,\u0000algorithms for cycle design, the conditions on which such algorithms are\u0000certain to work, as well as conditions for cycle stability remain elusive.\u0000Here, we state biologically motivated hypotheses that guarantee existence of\u0000such cycles in two deterministic classes of mathematical models already\u0000considered in the literature: Lotka-Volterra and adjusted replicator dynamics.\u0000We stress that not only existence of cyclic schemes, but also stability of such\u0000cycles is a relevant feature for applications in real clinical scenarios. We\u0000also analyze stochastic versions of the above deterministic models, a necessary\u0000step if we want to take into account that real tumors are composed by a finite\u0000population of cells subject to randomness, a relevant feature in the context of\u0000low tumor burden. We argue that the stability of the deterministic cycles is\u0000also relevant for the stochastic version of the models. In fact, Dua, Ma and\u0000Newton [Cancers (2021)] and Park and Newton [Phys. Rev. E (2023)] observed\u0000breakdown of deterministic cycles in a stochastic model (Moran process) for a\u0000tumor. Our findings indicate that the breakdown phenomenon is not due to\u0000stochasticity itself, but to the deterministic instability inherent in the\u0000cycles of the referenced papers. We then illustrate how stable deterministic\u0000cycles avoid for very large times the breakdown of cyclic treatments in\u0000stochastic tumor models.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"152 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Stathis Megas, Daniel G. Chen, Krzysztof Polanski, Moshe Eliasof, Carola-Bibiane Schonlieb, Sarah A. Teichmann
{"title":"Celcomen: spatial causal disentanglement for single-cell and tissue perturbation modeling","authors":"Stathis Megas, Daniel G. Chen, Krzysztof Polanski, Moshe Eliasof, Carola-Bibiane Schonlieb, Sarah A. Teichmann","doi":"arxiv-2409.05804","DOIUrl":"https://doi.org/arxiv-2409.05804","url":null,"abstract":"Celcomen leverages a mathematical causality framework to disentangle intra-\u0000and inter- cellular gene regulation programs in spatial transcriptomics and\u0000single-cell data through a generative graph neural network. It can learn\u0000gene-gene interactions, as well as generate post-perturbation counterfactual\u0000spatial transcriptomics, thereby offering access to experimentally inaccessible\u0000samples. We validated its disentanglement, identifiability, and counterfactual\u0000prediction capabilities through simulations and in clinically relevant human\u0000glioblastoma, human fetal spleen, and mouse lung cancer samples. Celcomen\u0000provides the means to model disease and therapy induced changes allowing for\u0000new insights into single-cell spatially resolved tissue responses relevant to\u0000human health.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"2 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Histopathological study on goldfish (Carassius auratus) gonads exposed to tobacco smoke","authors":"Ali Parsakhanghah","doi":"arxiv-2409.05175","DOIUrl":"https://doi.org/arxiv-2409.05175","url":null,"abstract":"Ornamental fish have various positive effects in human life. Due to the\u0000effect and importance of aesthetics and artificial reproduction of these fish,\u0000which may be enclosed in aquarium environments and come in contact with\u0000cigarette smoke, the effects of tobacco smoke on the gonad tissue of goldfish\u0000were investigated. For this purpose, 60 goldfish randomly (weight 100gr)\u0000divided in 3 groups and were released in tanks containing 10 liters of water\u0000(temperature: 20, hardness: 14 ppt, pH: 7/8). After adaptation, in treatment 1,\u00001gr of tobacco was heated daily with a direct flame, and the resulting smoke\u0000was collected and injected into water with an air pump, in treatment 2, this\u0000process was done twice a day. After 3 months, fish gonads tissue were sampled\u0000and histopathological slides were investigated. The results showed that in the\u0000treatment 2, there were early and immature oocytes in the ovarian tissue in\u0000comparison to other groups. Also, in the testes of fish of treatment 2, the\u0000reduction of spermatozoids and the higher number of spermatogonia were\u0000observed. In the treatment 3, these changes were more. A significant difference\u0000between the groups in both female and male was observed at the sexual\u0000maturation stages (P<0.001). Based on this study, the dissolution of tobacco\u0000smoke can have a negative effect on the process of sexual reproduction and fish\u0000exposed to more smoke are more likely to be sterile, and these changes were\u0000observed in both males and female.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"15 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Explainable AI for computational pathology identifies model limitations and tissue biomarkers","authors":"Jakub R. Kaczmarzyk, Joel H. Saltz, Peter K. Koo","doi":"arxiv-2409.03080","DOIUrl":"https://doi.org/arxiv-2409.03080","url":null,"abstract":"Deep learning models have shown promise in histopathology image analysis, but\u0000their opaque decision-making process poses challenges in high-risk medical\u0000scenarios. Here we introduce HIPPO, an explainable AI method that interrogates\u0000attention-based multiple instance learning (ABMIL) models in computational\u0000pathology by generating counterfactual examples through tissue patch\u0000modifications in whole slide images. Applying HIPPO to ABMIL models trained to\u0000detect breast cancer metastasis reveals that they may overlook small tumors and\u0000can be misled by non-tumor tissue, while attention maps$unicode{x2014}$widely\u0000used for interpretation$unicode{x2014}$often highlight regions that do not\u0000directly influence predictions. By interpreting ABMIL models trained on a\u0000prognostic prediction task, HIPPO identified tissue areas with stronger\u0000prognostic effects than high-attention regions, which sometimes showed\u0000counterintuitive influences on risk scores. These findings demonstrate HIPPO's\u0000capacity for comprehensive model evaluation, bias detection, and quantitative\u0000hypothesis testing. HIPPO greatly expands the capabilities of explainable AI\u0000tools to assess the trustworthy and reliable development, deployment, and\u0000regulation of weakly-supervised models in computational pathology.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"40 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Phase changes of the flow rate in the vertebral artery caused by debranching thoracic endovascular aortic repair: effects of flow path and local vessel stiffness on vertebral arterial pulsation","authors":"Naoki Takeishia, Li Jialongb, Naoto Yokoyamac, Hisashi Tanakad, Takasumi Gotoe, Shigeo Wada","doi":"arxiv-2409.02476","DOIUrl":"https://doi.org/arxiv-2409.02476","url":null,"abstract":"Despite numerous studies on cerebral arterial blood flow, there has not yet\u0000been a comprehensive description of hemodynamics in patients undergoing\u0000debranching thoracic endovascular aortic repair (dTEVAR), a promising surgical\u0000option for aortic arch aneurysms. A phase delay of the flow rate in the left\u0000vertebral artery (LVA) in patients after dTEVAR compared to those before was\u0000experimentally observed, while the phase in the right vertebral artery (RVA)\u0000remained almost the same before and after surgery. Since this surgical\u0000intervention included stent graft implantation and extra-anatomical bypass, it\u0000was expected that the intracranial hemodynamic changes due to dTEVAR were\u0000coupled with fluid flow and pulse waves in cerebral arteries. To clarify this\u0000issue, A one-dimensional model (1D) was used to numerically investigate the\u0000relative contribution (i.e., local vessel stiffness and flow path changes) of\u0000the VA flow rate to the phase difference. The numerical results demonstrated a\u0000phase delay of flow rate in the LVA but not the RVA in postoperative patients\u0000undergoing dTEVAR relative to preoperative patients. The results further showed\u0000that the primary factor affecting the phase delay of the flow rate in the LVA\u0000after surgery compared to that before was the bypass, i.e., alteration of flow\u0000path, rather than stent grafting, i.e., the change in local vessel stiffness.\u0000The numerical results provide insights into hemodynamics in postoperative\u0000patients undergoing dTEVAR, as well as knowledge about therapeutic decisions.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"388 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jixin Hou, Zhengwang Wu, Xianyan Chen, Dajiang Zhu, Tianming Liu, Gang Li, Xianqiao Wang
{"title":"Role of Data-driven Regional Growth Model in Shaping Brain Folding Patterns","authors":"Jixin Hou, Zhengwang Wu, Xianyan Chen, Dajiang Zhu, Tianming Liu, Gang Li, Xianqiao Wang","doi":"arxiv-2408.17334","DOIUrl":"https://doi.org/arxiv-2408.17334","url":null,"abstract":"The surface morphology of the developing mammalian brain is crucial for\u0000understanding brain function and dysfunction. Computational modeling offers\u0000valuable insights into the underlying mechanisms for early brain folding. While\u0000previous studies generally assume uniform growth, recent findings indicate\u0000significant regional variations in brain tissue growth. However, the role of\u0000these variations in cortical development remains unclear. In this study, we\u0000explored how regional cortical growth affects brain folding patterns. We first\u0000developed growth models for typical cortical regions using ML-assisted symbolic\u0000regression, based on longitudinal data from over 1,000 infant MRI scans that\u0000captured cortical surface area and thickness during perinatal and postnatal\u0000brains development. These models were subsequently integrated into\u0000computational software to simulate cortical development with anatomically\u0000realistic geometric models. We quantified the resulting folding patterns using\u0000metrics such as mean curvature, sulcal depth, and gyrification index. Our\u0000results demonstrate that regional growth models generate complex brain folding\u0000patterns that more closely match actual brains structures, both quantitatively\u0000and qualitatively, compared to uniform growth models. Growth magnitude plays a\u0000dominant role in shaping folding patterns, while growth trajectory has a minor\u0000influence. Moreover, multi-region models better capture the intricacies of\u0000brain folding than single-region models. Our results underscore the necessity\u0000and importance of incorporating regional growth heterogeneity into brain\u0000folding simulations, which could enhance early diagnosis and treatment of\u0000cortical malformations and neurodevelopmental disorders such as epilepsy and\u0000autism.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Programmable scanning diffuse speckle contrast imaging of cerebral blood flow","authors":"Faezeh Akbari, Xuhui Liu, Fatemeh Hamedi, Mehrana Mohtasebi, Lei Chen, Guoqiang Yu","doi":"arxiv-2408.12715","DOIUrl":"https://doi.org/arxiv-2408.12715","url":null,"abstract":"Significance: Cerebral blood flow (CBF) imaging is crucial for diagnosing\u0000cerebrovascular diseases. However, existing large neuroimaging techniques with\u0000high cost, low sampling rate, and poor mobility make them unsuitable for\u0000continuous and longitudinal CBF monitoring at the bedside. Aim: This study\u0000aimed to develop a low-cost, portable, programmable scanning diffuse speckle\u0000contrast imaging (PS-DSCI) technology for fast, high-density, and\u0000depth-sensitive imaging of CBF in rodents. Approach: The PS-DSCI employed a\u0000programmable digital micromirror device (DMD) for remote line-shape laser (785\u0000nm) scanning on tissue surface and synchronized a 2D camera for capturing\u0000boundary diffuse laser speckle contrasts. New algorithms were developed to\u0000address deformations of line-shape scanning, thus minimizing CBF reconstruction\u0000artifacts. The PS-DSCI was examined in head-simulating phantoms and adult mice.\u0000Results: The PS-DSCI enables resolving Intralipid particle flow contrasts at\u0000different tissue depths. In vivo experiments in adult mice demonstrated the\u0000capability of PS-DSCI to image global/regional CBF variations induced by 8% CO2\u0000inhalation and transient carotid artery ligations. Conclusions: Compared to\u0000conventional point scanning, the line scanning in PS-DSCI significantly\u0000increases spatiotemporal resolution. The high sampling rate of PS-DSCI is\u0000crucial for capturing rapid CBF changes while high spatial resolution is\u0000important for visualizing brain vasculature.","PeriodicalId":501572,"journal":{"name":"arXiv - QuanBio - Tissues and Organs","volume":"56 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}