A deep learning-based pipeline for analyzing the influences of interfacial mechanochemical microenvironments on spheroid invasion using differential interference contrast microscopic images

Metastasis is the leading cause of cancer-related deaths. During this process, cancer cells are likely to navigate discrete tissue-tissue interfaces, enabling them to infiltrate and spread throughout the body. Three-dimensional (3D) spheroid modeling is receiving more attention due to its strengths in studying the invasive behavior of metastatic cancer cells. While microscopy is a conventional approach for investigating 3D invasion, post-invasion image analysis, which is a time-consuming process, remains a significant challenge for researchers. In this study, we presented an image processing pipeline that utilized a deep learning (DL) solution, with an encoder-decoder architecture, to assess and characterize the invasion dynamics of tumor spheroids. The developed models, equipped with feature extraction and measurement capabilities, could be successfully utilized for the automated segmentation of the invasive protrusions as well as the core region of spheroids situated within interfacial microenvironments with distinct mechanochemical factors. Our findings suggest that a combination of the spheroid culture and DL-based image analysis enable identification of time-lapse migratory patterns for tumor spheroids above matrix-substrate interfaces, thus paving the foundation for delineating the mechanism of local invasion during cancer metastasis.

DCNNLFS: A Dilated Convolutional Neural Network with Late Fusion Strategy for Intelligent Classification of Gastric Histopathology Images

Gastric cancer has a high incidence rate, significantly threatening patients' health. Gastric histopathology images can reliably diagnose related diseases. Still, the data volume of histopathology images is too large, making misdiagnosis or missed diagnosis easy. The classification model based on deep learning has made some progress on gastric histopathology images. However, traditional convolutional neural networks (CNN) generally use pooling operations, which will reduce the spatial resolution of the image, resulting in poor prediction results. The image feature in previous CNN has a poor perception of details. Therefore, we design a dilated CNN with a late fusion strategy (DCNNLFS) for gastric histopathology image classification. The DCNNLFS model utilizes dilated convolutions, enabling it to expand the receptive field. The dilated convolutions can learn the different contextual information by adjusting the dilation rate. The DCNNLFS model uses a late fusion strategy to enhance the classification ability of DCNNLFS. We run related experiments on a gastric histopathology image dataset to verify the excellence of the DCNNLFS model, where the three metrics Precision, Accuracy, and F1-Score are 0.938, 0.935, and 0.959.

The accuracy of international and national fetal growth charts in detecting small-for-gestational-age infants using the Lambda-Mu-Sigma method

Objective

To construct a national fetal growth chart using retrospective data and compared its diagnostic accuracy in predicting SGA at birth with existing international growth charts.

Method

This is a retrospective study where datasets from May 2011 to Apr 2020 were extracted to construct the fetal growth chart using the Lambda-Mu-Sigma method. SGA is defined as birth weight <10th centile. The local growth chart's diagnostic accuracy in detecting SGA at birth was evaluated using datasets from May 2020 to Apr 2021 and was compared with the WHO, Hadlock, and INTERGROWTH-21st charts. Balanced accuracy, sensitivity, and specificity were reported.

Results

A total of 68,897 scans were collected and five biometric growth charts were constructed. Our national growth chart achieved an accuracy of 69% and a sensitivity of 42% in identifying SGA at birth. The WHO chart showed similar diagnostic performance as our national growth chart, followed by the Hadlock (67% accuracy and 38% sensitivity) and INTERGROWTH-21st (57% accuracy and 19% sensitivity). The specificities for all charts were 95-96%. All growth charts showed higher accuracy in the third trimester, with an improvement of 8-16%, as compared to that in the second trimester.

Conclusion

Using the Hadlock and INTERGROWTH-21st chart in the Malaysian population may results in misdiagnose of SGA. Our population local chart has slightly higher accuracy in predicting preterm SGA in the second trimester which can enable earlier intervention for babies who are detected as SGA. All growth charts' diagnostic accuracies were poor in the second trimester, suggesting the need of improvising alternative techniques for early detection of SGA to improve fetus outcomes.

A Miniaturized Ultrasonic Micro-hole Perforator for Minimally Invasive Craniotomy

OBJECTIVE

Micro-hole perforation on skull is urgently desired for minimally invasive insertion of micro-tools in brain for diagnostic or treatment purpose. However, a micro drill bit would easily fracture, making it difficult to safely generate a micro-hole on the hard skull.

METHODS

In this study, we present a method for ultrasonic vibration assisted micro-hole perforation on skull in a manner similar to subcutaneous injection on soft tissue. For this purpose, a high amplitude miniaturized ultrasonic tool with a 500 μm tip diameter micro-hole perforator was developed with simulation and experimental characterization. In-depth investigation of micro-hole generation mechanism was performed with systematic experiments on animal skull with a bespoke test rig; effects of vibration amplitude and feed rate on hole forming characteristics were systematically studied. It was observed that by exploiting skull bone's unique structural and material properties, the ultrasonic micro-perforator could locally damage bone tissue with micro-porosities, induce sufficient plastic deformation to bone tissue around the micro-hole and refrain elastic recovery after tool withdraw, generating a micro-hole on skull without material.

RESULTS

Under optimized conditions, high quality micro-holes could be formed on the hard skull with a force (< 1N) even smaller than that for subcutaneous injection on soft skin.

CONCLUSION

This study would provide a safe and effective method and a miniaturized device for micro-hole perforation on skull for minimally invasive neural interventions.

Current challenges of implementing artificial intelligence in medical imaging

The idea of using artificial intelligence (AI) in medical practice has gained vast interest due to its potential to revolutionise healthcare systems. However, only some AI algorithms are utilised due to systems' uncertainties, besides the never-ending list of ethical and legal concerns. This paper intends to provide an overview of current AI challenges in medical imaging with an ultimate aim to foster better and effective communication among various stakeholders to encourage AI technology development. We identify four main challenges in implementing AI in medical imaging, supported with consequences and past events when these problems fail to mitigate. Among them is the creation of a robust AI algorithm that is fair, trustable and transparent. Another issue is on data governance, in which best practices in data sharing must be established to promote trust and protect the patients' privacy. Next, stakeholders, such as the government, technology companies and hospital management, should come to a consensus in creating trustworthy AI policies and regulatory frameworks, which is the fourth challenge, to support, encourage and spur innovation in digital AI healthcare technology. Lastly, we discussed the efforts of various organizations such as the World Health Organisation (WHO), American College of Radiology (ACR), European Society of Radiology (ESR) and Radiological Society of North America (RSNA), who are already actively pursuing ethical developments in AI. The efforts by various stakeholders will eventually overcome hurdles and the deployment of AI-driven healthcare applications in clinical practice will become a reality and hence lead to better healthcare services and outcomes.

Interpreting the role of nuchal fold for fetal growth restriction prediction using machine learning

The objective of the study is to investigate the effect of Nuchal Fold (NF) in predicting Fetal Growth Restriction (FGR) using machine learning (ML), to explain the model's results using model-agnostic interpretable techniques, and to compare the results with clinical guidelines. This study used second-trimester ultrasound biometry and Doppler velocimetry were used to construct six FGR (birthweight < 3rd centile) ML models. Interpretability analysis was conducted using Accumulated Local Effects (ALE) and Shapley Additive Explanations (SHAP). The results were compared with clinical guidelines based on the most optimal model. Support Vector Machine (SVM) exhibited the most consistent performance in FGR prediction. SHAP showed that the top contributors to identify FGR were Abdominal Circumference (AC), NF, Uterine RI (Ut RI), and Uterine PI (Ut PI). ALE showed that the cutoff values of Ut RI, Ut PI, and AC in differentiating FGR from normal were comparable with clinical guidelines (Errors between model and clinical; Ut RI: 15%, Ut PI: 8%, and AC: 11%). The cutoff value for NF to differentiate between healthy and FGR is 5.4 mm, where low NF may indicate FGR. The SVM model is the most stable in FGR prediction. ALE can be a potential tool to identify a cutoff value for novel parameters to differentiate between healthy and FGR.

A Review of Biomechanics Analysis of the Umbilical-Placenta System With Regards to Diseases

Placenta is an important organ that is crucial for both fetal and maternal health. Abnormalities of the placenta, such as during intrauterine growth restriction (IUGR) and pre-eclampsia (PE) are common, and an improved understanding of these diseases is needed to improve medical care. Biomechanics analysis of the placenta is an under-explored area of investigation, which has demonstrated usefulness in contributing to our understanding of the placenta physiology. In this review, we introduce fundamental biomechanics concepts and discuss the findings of biomechanical analysis of the placenta and umbilical cord, including both tissue biomechanics and biofluid mechanics. The biomechanics of placenta ultrasound elastography and its potential in improving clinical detection of placenta diseases are also discussed. Finally, potential future work is listed.

Model learning analysis of 3D optoacoustic mesoscopy images for the classification of atopic dermatitis

Atopic dermatitis (AD) is a skin inflammatory disease affecting 10% of the population worldwide. Raster-scanning optoacoustic mesoscopy (RSOM) has recently shown promise in dermatological imaging. We conducted a comprehensive analysis using three machine-learning models, random forest (RF), support vector machine (SVM), and convolutional neural network (CNN) for classifying healthy versus AD conditions, and sub-classifying different AD severities using RSOM images and clinical information. CNN model successfully differentiates healthy from AD patients with 97% accuracy. With limited data, RF achieved 65% accuracy in sub-classifying AD patients into mild versus moderate-severe cases. Identification of disease severities is vital in managing AD treatment.

Machine learning improves early prediction of small-for-gestational-age births and reveals nuchal fold thickness as unexpected predictor

OBJECTIVE

To investigate the performance of the machine learning (ML) model in predicting small-for-gestational-age (SGA) at birth, using second-trimester data.

METHODS

Retrospective data of 347 patients, consisting of maternal demographics and ultrasound parameters collected between the 20th and 25th gestational weeks, were studied. ML models were applied to different combinations of the parameters to predict SGA and severe SGA at birth (defined as 10th and third centile birth weight).

RESULTS

Using second-trimester measurements, ML models achieved an accuracy of 70% and 73% in predicting SGA and severe SGA whereas clinical guidelines had accuracies of 64% and 48%. Uterine PI (Ut PI) was found to be an important predictor, corroborating with existing literature, but surprisingly, so was nuchal fold thickness (NF). Logistic regression showed that Ut PI and NF were significant predictors and statistical comparisons showed that these parameters were significantly different in disease. Further, including NF was found to improve ML model performance, and vice versa.

CONCLUSION

ML could potentially improve the prediction of SGA at birth from second-trimester measurements, and demonstrated reduced NF to be an important predictor. Early prediction of SGA allows closer clinical monitoring, which provides an opportunity to discover any underlying diseases associated with SGA.

Differences in placental capillary shear stress in fetal growth restriction may affect endothelial cell function and vascular network formation

Fetal growth restriction (FGR) affects 5-10% of pregnancies, leading to clinically significant fetal morbidity and mortality. FGR placentae frequently exhibit poor vascular branching, but the mechanisms driving this are poorly understood. We hypothesize that vascular structural malformation at the organ level alters microvascular shear stress, impairing angiogenesis. A computational model of placental vasculature predicted elevated placental micro-vascular shear stress in FGR placentae (0.2 Pa in severe FGR vs 0.05 Pa in normal placentae). Endothelial cells cultured under predicted FGR shear stresses migrated significantly slower and with greater persistence than in shear stresses predicted in normal placentae. These cell behaviors suggest a dominance of vessel elongation over branching. Taken together, these results suggest (1) poor vascular development increases vessel shear stress, (2) increased shear stress induces cell behaviors that impair capillary branching angiogenesis, and (3) impaired branching angiogenesis continues to drive elevated shear stress, jeopardizing further vascular formation. Inadequate vascular branching early in gestation could kick off this cyclic loop and continue to negatively impact placental angiogenesis throughout gestation.

Altered Placental Chorionic Arterial Biomechanical Properties During Intrauterine Growth Restriction

Intrauterine growth restriction (IUGR) is a pregnancy complication due to placental dysfunction that prevents the fetus from obtaining enough oxygen and nutrients, leading to serious mortality and morbidity risks. There is no treatment for IUGR despite having a prevalence of 3% in developed countries, giving rise to an urgency to improve our understanding of the disease. Applying biomechanics investigation on IUGR placental tissues can give important new insights. We performed pressure-diameter mechanical testing of placental chorionic arteries and found that in severe IUGR cases (RI > 90th centile) but not in IUGR cases (RI < 90th centile), vascular distensibility was significantly increased from normal. Constitutive modeling demonstrated that a simplified Fung-type hyperelastic model was able to describe the mechanical properties well, and histology showed that severe IUGR had the lowest collagen to elastin ratio. To demonstrate that the increased distensibility in the severe IUGR group was related to their elevated umbilical resistance and pulsatility indices, we modelled the placental circulation using a Windkessel model, and demonstrated that vascular compliance (and not just vascular resistance) directly affected blood flow pulsatility, suggesting that it is an important parameter for the disease. Our study showed that biomechanics study on placenta could extend our understanding on placenta physiology.

Characterization of the hemodynamic wall shear stresses in human umbilical vessels from normal and intrauterine growth restricted pregnancies

Significant reductions in blood flow and umbilical diameters were reported in pregnancies affected by intrauterine growth restriction (IUGR) from placental insufficiency. However, it is not known if IUGR umbilical blood vessels experience different hemodynamic wall shear stresses (WSS) compared to normal umbilical vessels. As WSS is known to influence vasoactivity and vascular growth and remodeling, which can regulate flow rates, it is important to study this parameter. In this study, we aim to characterize umbilical vascular WSS environment in normal and IUGR pregnancies, and evaluate correlation between WSS and vascular diameter, and gestational age. Twenty-two normal and 21 IUGR pregnancies were assessed via ultrasound between the 27th and 39th gestational week. IUGR was defined as estimated fetal weight and/or abdominal circumference below the 10th centile, with no improvement during the remainder of the pregnancy. Vascular diameter was determined by 3D ultrasound scans and image segmentation. Umbilical artery (UA) WSS was computed via computational flow simulations, while umbilical vein (UV) WSS was computed via the Poiseuille equation. Univariate multiple regression analysis was used to test for the differences between normal and IUGR cohort. UV volumetric flow rate, UA and UV diameters were significantly lower in IUGR fetuses, but flow velocities and WSS trends in UA and UV were very similar between normal and IUGR groups. In both groups, UV WSS showed a significant negative correlation with diameter, but UA WSS had no correlation with diameter, suggesting a constancy of WSS environment and the existence of WSS homeostasis in UA, but not in UV. Despite having reduced flow rate and vascular sizes, IUGR UAs had hemodynamic mechanical stress environments and trends that were similar to those in normal pregnancies. This suggested that endothelial dysfunction or abnormal mechanosensing was unlikely to be the cause of small vessels in IUGR umbilical cords.

Motorizing and Optimizing Ultrasound Strain Elastography for Detection of Intrauterine Growth Restriction Pregnancies

Intrauterine growth restriction is a prevalent disease in pregnancy in which placental insufficiency leads to 5 to 10 times higher mortality and lifelong morbidities. The current detection rate is poor, and recently, ultrasound strain elastography (USEL) was proposed as a new diagnostic technique. Currently, placental USEL uses maternal subcutaneous fat as the reference layer, but this is not ideal as fat tissue stiffness can vary widely between subjects. Current USEL also uses manual palpation, and under different compression depths and rates, viscoelastic tissues such as placenta can yield different stiffness results. In the study described here, we strove to improve placental USEL by (i) using an external polymeric pad of known stiffness as the reference layer and (ii) adopting motorized control of the transducer during USEL to standardize palpation motion. Results indicated that motorized USEL reduced measurement variability by 67% compared with freehand USEL. Satisfactory and statistically significant correlations between USEL measurements and mechanical testing validation results were obtained for our new USEL protocol. Placental tissues were found to be non-linear and viscoelastic in nature and, thus, differed in stiffness at different compression rates and depths. Our study also revealed that there was a specific compression depth and rate during USEL that provided better correlation to mechanical testing, and should be considered in clinical placental USEL.