GO2Sum: generating human-readable functional summary of proteins from GO terms.

Understanding the biological functions of proteins is of fundamental importance in modern biology. To represent a function of proteins, Gene Ontology (GO), a controlled vocabulary, is frequently used, because it is easy to handle by computer programs avoiding open-ended text interpretation. Particularly, the majority of current protein function prediction methods rely on GO terms. However, the extensive list of GO terms that describe a protein function can pose challenges for biologists when it comes to interpretation. In response to this issue, we developed GO2Sum (Gene Ontology terms Summarizer), a model that takes a set of GO terms as input and generates a human-readable summary using the T5 large language model. GO2Sum was developed by fine-tuning T5 on GO term assignments and free-text function descriptions for UniProt entries, enabling it to recreate function descriptions by concatenating GO term descriptions. Our results demonstrated that GO2Sum significantly outperforms the original T5 model that was trained on the entire web corpus in generating Function, Subunit Structure, and Pathway paragraphs for UniProt entries.

Domain-PFP allows protein function prediction using function-aware domain embedding representations.

Domains are functional and structural units of proteins that govern various biological functions performed by the proteins. Therefore, the characterization of domains in a protein can serve as a proper functional representation of proteins. Here, we employ a self-supervised protocol to derive functionally consistent representations for domains by learning domain-Gene Ontology (GO) co-occurrences and associations. The domain embeddings we constructed turned out to be effective in performing actual function prediction tasks. Extensive evaluations showed that protein representations using the domain embeddings are superior to those of large-scale protein language models in GO prediction tasks. Moreover, the new function prediction method built on the domain embeddings, named Domain-PFP, substantially outperformed the state-of-the-art function predictors. Additionally, Domain-PFP demonstrated competitive performance in the CAFA3 evaluation, achieving overall the best performance among the top teams that participated in the assessment.

GO2Sum: Generating Human Readable Functional Summary of Proteins from GO Terms.

Understanding the biological functions of proteins is of fundamental importance in modern biology. To represent function of proteins, Gene Ontology (GO), a controlled vocabulary, is frequently used, because it is easy to handle by computer programs avoiding open-ended text interpretation. Particularly, the majority of current protein function prediction methods rely on GO terms. However, the extensive list of GO terms that describe a protein function can pose challenges for biologists when it comes to interpretation. In response to this issue, we developed GO2Sum (Gene Ontology terms Summarizer), a model that takes a set of GO terms as input and generates a human-readable summary using the T5 large language model. GO2Sum was developed by fine-tuning T5 on GO term assignments and free-text function descriptions for UniProt entries, enabling it to recreate function descriptions by concatenating GO term descriptions. Our results demonstrated that GO2Sum significantly outperforms the original T5 model that was trained on the entire web corpus in generating Function, Subunit Structure, and Pathway paragraphs for UniProt entries.

NuFold: A Novel Tertiary RNA Structure Prediction Method Using Deep Learning with Flexible Nucleobase Center Representation.

RNA is not only playing a core role in the central dogma as mRNA between DNA and protein, but also many non-coding RNAs have been discovered to have unique and diverse biological functions. As genome sequences become increasingly available and our knowledge of RNA sequences grows, the study of RNA's structure and function has become more demanding. However, experimental determination of three-dimensional RNA structures is both costly and time-consuming, resulting in a substantial disparity between RNA sequence data and structural insights. In response to this challenge, we propose a novel computational approach that harnesses state-of-the-art deep learning architecture NuFold to accurately predict RNA tertiary structures. This approach aims to offer a cost-effective and efficient means of bridging the gap between RNA sequence information and structural comprehension. NuFold implements a nucleobase center representation, which allows it to reproduce all possible nucleotide conformations accurately.

Domain-PFP: Protein Function Prediction Using Function-Aware Domain Embedding Representations.

Domains are functional and structural units of proteins that govern various biological functions performed by the proteins. Therefore, the characterization of domains in a protein can serve as a proper functional representation of proteins. Here, we employ a self-supervised protocol to derive functionally consistent representations for domains by learning domain-Gene Ontology (GO) co-occurrences and associations. The domain embeddings we constructed turned out to be effective in performing actual function prediction tasks. Extensive evaluations showed that protein representations using the domain embeddings are superior to those of large-scale protein language models in GO prediction tasks. Moreover, the new function prediction method built on the domain embeddings, named Domain-PFP, significantly outperformed the state-of-the-art function predictors. Additionally, Domain-PFP demonstrated competitive performance in the CAFA3 evaluation, achieving overall the best performance among the top teams that participated in the assessment.

Align-gram: Rethinking the Skip-gram Model for Protein Sequence Analysis.

The inception of next generations sequencing technologies have exponentially increased the volume of biological sequence data. Protein sequences, being quoted as the 'language of life', has been analyzed for a multitude of applications and inferences. Owing to the rapid development of deep learning, in recent years there have been a number of breakthroughs in the domain of Natural Language Processing. Since these methods are capable of performing different tasks when trained with a sufficient amount of data, off-the-shelf models are used to perform various biological applications. In this study, we investigated the applicability of the popular Skip-gram model for protein sequence analysis and made an attempt to incorporate some biological insights into it. We propose a novel k-mer embedding scheme, Align-gram, which is capable of mapping the similar k-mers close to each other in a vector space. Furthermore, we experiment with other sequence-based protein representations and observe that the embeddings derived from Align-gram aids modeling and training deep learning models better. Our experiments with a simple baseline LSTM model and a much complex CNN model of DeepGoPlus shows the potential of Align-gram in performing different types of deep learning applications for protein sequence analysis.

PPG2ABP: Translating Photoplethysmogram (PPG) Signals to Arterial Blood Pressure (ABP) Waveforms.

Cardiovascular diseases are one of the most severe causes of mortality, annually taking a heavy toll on lives worldwide. Continuous monitoring of blood pressure seems to be the most viable option, but this demands an invasive process, introducing several layers of complexities and reliability concerns due to non-invasive techniques not being accurate. This motivates us to develop a method to estimate the continuous arterial blood pressure (ABP) waveform through a non-invasive approach using Photoplethysmogram (PPG) signals. We explore the advantage of deep learning, as it would free us from sticking to ideally shaped PPG signals only by making handcrafted feature computation irrelevant, which is a shortcoming of the existing approaches. Thus, we present PPG2ABP, a two-stage cascaded deep learning-based method that manages to estimate the continuous ABP waveform from the input PPG signal with a mean absolute error of 4.604 mmHg, preserving the shape, magnitude, and phase in unison. However, the more astounding success of PPG2ABP turns out to be that the computed values of Diastolic Blood Pressure (DBP), Mean Arterial Pressure (MAP), and Systolic Blood Pressure (SBP) from the estimated ABP waveform outperform the existing works under several metrics (mean absolute error of 3.449 ± 6.147 mmHg, 2.310 ± 4.437 mmHg, and 5.727 ± 9.162 mmHg, respectively), despite that PPG2ABP is not explicitly trained to do so. Notably, both for DBP and MAP, we achieve Grade A in the BHS (British Hypertension Society) Standard and satisfy the AAMI (Association for the Advancement of Medical Instrumentation) standard.

Differentiation and classification of bacterial endotoxins based on surface enhanced Raman scattering and advanced machine learning.

Bacterial endotoxin, a major component of the Gram-negative bacterial outer membrane leaflet, is a lipopolysaccharide shed from bacteria during their growth and infection and can be utilized as a biomarker for bacterial detection. Here, the surface enhanced Raman scattering (SERS) spectra of eleven bacterial endotoxins with an average detection amount of 8.75 pg per measurement have been obtained based on silver nanorod array substrates, and the characteristic SERS peaks have been identified. With appropriate spectral pre-processing procedures, different classical machine learning algorithms, including support vector machine, k-nearest neighbor, random forest, etc., and a modified deep learning algorithm, RamanNet, have been applied to differentiate and classify these endotoxins. It has been found that most conventional machine learning algorithms can attain a differentiation accuracy of >99%, while RamanNet can achieve 100% accuracy. Such an approach has the potential for precise classification of endotoxins and could be used for rapid medical diagnoses and therapeutic decisions for pathogenic infections.

QUCoughScope: An Intelligent Application to Detect COVID-19 Patients Using Cough and Breath Sounds.

Problem-Since the outbreak of the COVID-19 pandemic, mass testing has become essential to reduce the spread of the virus. Several recent studies suggest that a significant number of COVID-19 patients display no physical symptoms whatsoever. Therefore, it is unlikely that these patients will undergo COVID-19 testing, which increases their chances of unintentionally spreading the virus. Currently, the primary diagnostic tool to detect COVID-19 is a reverse-transcription polymerase chain reaction (RT-PCR) test from the respiratory specimens of the suspected patient, which is invasive and a resource-dependent technique. It is evident from recent researches that asymptomatic COVID-19 patients cough and breathe in a different way than healthy people. Aim-This paper aims to use a novel machine learning approach to detect COVID-19 (symptomatic and asymptomatic) patients from the convenience of their homes so that they do not overburden the healthcare system and also do not spread the virus unknowingly by continuously monitoring themselves. Method-A Cambridge University research group shared such a dataset of cough and breath sound samples from 582 healthy and 141 COVID-19 patients. Among the COVID-19 patients, 87 were asymptomatic while 54 were symptomatic (had a dry or wet cough). In addition to the available dataset, the proposed work deployed a real-time deep learning-based backend server with a web application to crowdsource cough and breath datasets and also screen for COVID-19 infection from the comfort of the user's home. The collected dataset includes data from 245 healthy individuals and 78 asymptomatic and 18 symptomatic COVID-19 patients. Users can simply use the application from any web browser without installation and enter their symptoms, record audio clips of their cough and breath sounds, and upload the data anonymously. Two different pipelines for screening were developed based on the symptoms reported by the users: asymptomatic and symptomatic. An innovative and novel stacking CNN model was developed using three base learners from of eight state-of-the-art deep learning CNN algorithms. The stacking CNN model is based on a logistic regression classifier meta-learner that uses the spectrograms generated from the breath and cough sounds of symptomatic and asymptomatic patients as input using the combined (Cambridge and collected) dataset. Results-The stacking model outperformed the other eight CNN networks with the best classification performance for binary classification using cough sound spectrogram images. The accuracy, sensitivity, and specificity for symptomatic and asymptomatic patients were 96.5%, 96.42%, and 95.47% and 98.85%, 97.01%, and 99.6%, respectively. For breath sound spectrogram images, the metrics for binary classification of symptomatic and asymptomatic patients were 91.03%, 88.9%, and 91.5% and 80.01%, 72.04%, and 82.67%, respectively. Conclusion-The web-application QUCoughScope records coughing and breathing sounds, converts them to a spectrogram, and applies the best-performing machine learning model to classify the COVID-19 patients and healthy subjects. The result is then reported back to the test user in the application interface. Therefore, this novel system can be used by patients in their premises as a pre-screening method to aid COVID-19 diagnosis by prioritizing the patients for RT-PCR testing and thereby reducing the risk of spreading of the disease.

Revisiting segmentation of lung tumors from CT images.

Lung cancer is a leading cause of death throughout the world. Because the prompt diagnosis of tumors allows oncologists to discern their nature, type, and mode of treatment, tumor detection and segmentation from CT scan images is a crucial field of study. This paper investigates lung tumor segmentation via a two-dimensional Discrete Wavelet Transform (DWT) on the LOTUS dataset (31,247 training, and 4458 testing samples) and a Deeply Supervised MultiResUNet model. Coupling the DWT, which is used to achieve a more meticulous textural analysis while integrating information from neighboring CT slices, with the deep supervision of the model architecture results in an improved dice coefficient of 0.8472. A key characteristic of our approach is its avoidance of 3D kernels (despite being used for a 3D segmentation task), thereby making it quite lightweight.

A Shallow U-Net Architecture for Reliably Predicting Blood Pressure (BP) from Photoplethysmogram (PPG) and Electrocardiogram (ECG) Signals.

Cardiovascular diseases are the most common causes of death around the world. To detect and treat heart-related diseases, continuous blood pressure (BP) monitoring along with many other parameters are required. Several invasive and non-invasive methods have been developed for this purpose. Most existing methods used in hospitals for continuous monitoring of BP are invasive. On the contrary, cuff-based BP monitoring methods, which can predict systolic blood pressure (SBP) and diastolic blood pressure (DBP), cannot be used for continuous monitoring. Several studies attempted to predict BP from non-invasively collectible signals such as photoplethysmograms (PPG) and electrocardiograms (ECG), which can be used for continuous monitoring. In this study, we explored the applicability of autoencoders in predicting BP from PPG and ECG signals. The investigation was carried out on 12,000 instances of 942 patients of the MIMIC-II dataset, and it was found that a very shallow, one-dimensional autoencoder can extract the relevant features to predict the SBP and DBP with state-of-the-art performance on a very large dataset. An independent test set from a portion of the MIMIC-II dataset provided a mean absolute error (MAE) of 2.333 and 0.713 for SBP and DBP, respectively. On an external dataset of 40 subjects, the model trained on the MIMIC-II dataset provided an MAE of 2.728 and 1.166 for SBP and DBP, respectively. For both the cases, the results met British Hypertension Society (BHS) Grade A and surpassed the studies from the current literature.

Robust biometric system using session invariant multimodal EEG and keystroke dynamics by the ensemble of self-ONNs.

Harnessing the inherent anti-spoofing quality from electroencephalogram (EEG) signals has become a potential field of research in recent years. Although several studies have been conducted, still there are some vital challenges present in the deployment of EEG-based biometrics, which is stable and capable of handling the real-world scenario. One of the key challenges is the large signal variability of EEG when recorded on different days or sessions which impedes the performance of biometric systems significantly. To address this issue, a session invariant multimodal Self-organized Operational Neural Network (Self-ONN) based ensemble model combining EEG and keystroke dynamics is proposed in this paper. Our model is tested successfully on a large number of sessions (10 recording days) with many challenging noisy and variable environments for the identification and authentication tasks. In most of the previous studies, training and testing were performed either over a single recording session (same day) only or without ensuring appropriate splitting of the data on multiple recording days. Unlike those studies, in our work, we have rigorously split the data so that train and test sets do not share the data of the same recording day. The proposed multimodal Self-ONN based ensemble model has achieved identification accuracy of 98% in rigorous validation cases and outperformed the equivalent ensemble of deep CNN models. A novel Self-ONN Siamese network has also been proposed to measure the similarity of templates during the authentication task instead of the commonly used simple distance measure techniques. The multimodal Siamese network reduces the Equal Error Rate (EER) to 1.56% in rigorous authentication. The obtained results indicate that the proposed multimodal Self-ONN model can automatically extract session invariant unique non-linear features to identify and authenticate users with high accuracy.

COVID-19 infection localization and severity grading from chest X-ray images.

The immense spread of coronavirus disease 2019 (COVID-19) has left healthcare systems incapable to diagnose and test patients at the required rate. Given the effects of COVID-19 on pulmonary tissues, chest radiographic imaging has become a necessity for screening and monitoring the disease. Numerous studies have proposed Deep Learning approaches for the automatic diagnosis of COVID-19. Although these methods achieved outstanding performance in detection, they have used limited chest X-ray (CXR) repositories for evaluation, usually with a few hundred COVID-19 CXR images only. Thus, such data scarcity prevents reliable evaluation of Deep Learning models with the potential of overfitting. In addition, most studies showed no or limited capability in infection localization and severity grading of COVID-19 pneumonia. In this study, we address this urgent need by proposing a systematic and unified approach for lung segmentation and COVID-19 localization with infection quantification from CXR images. To accomplish this, we have constructed the largest benchmark dataset with 33,920 CXR images, including 11,956 COVID-19 samples, where the annotation of ground-truth lung segmentation masks is performed on CXRs by an elegant human-machine collaborative approach. An extensive set of experiments was performed using the state-of-the-art segmentation networks, U-Net, U-Net++, and Feature Pyramid Networks (FPN). The developed network, after an iterative process, reached a superior performance for lung region segmentation with Intersection over Union (IoU) of 96.11% and Dice Similarity Coefficient (DSC) of 97.99%. Furthermore, COVID-19 infections of various shapes and types were reliably localized with 83.05% IoU and 88.21% DSC. Finally, the proposed approach has achieved an outstanding COVID-19 detection performance with both sensitivity and specificity values above 99%.

Development and Validation of an Early Scoring System for Prediction of Disease Severity in COVID-19 Using Complete Blood Count Parameters.

The coronavirus disease 2019 (COVID-19) after outbreaking in Wuhan increasingly spread throughout the world. Fast, reliable, and easily accessible clinical assessment of the severity of the disease can help in allocating and prioritizing resources to reduce mortality. The objective of the study was to develop and validate an early scoring tool to stratify the risk of death using readily available complete blood count (CBC) biomarkers. A retrospective study was conducted on twenty-three CBC blood biomarkers for predicting disease mortality for 375 COVID-19 patients admitted to Tongji Hospital, China from January 10 to February 18, 2020. Machine learning based key biomarkers among the CBC parameters as the mortality predictors were identified. A multivariate logistic regression-based nomogram and a scoring system was developed to categorize the patients in three risk groups (low, moderate, and high) for predicting the mortality risk among COVID-19 patients. Lymphocyte count, neutrophils count, age, white blood cell count, monocytes (%), platelet count, red blood cell distribution width parameters collected at hospital admission were selected as important biomarkers for death prediction using random forest feature selection technique. A CBC score was devised for calculating the death probability of the patients and was used to categorize the patients into three sub-risk groups: low (<=5%), moderate (>5% and <=50%), and high (>50%), respectively. The area under the curve (AUC) of the model for the development and internal validation cohort were 0.961 and 0.88, respectively. The proposed model was further validated with an external cohort of 103 patients of Dhaka Medical College, Bangladesh, which exhibits in an AUC of 0.963. The proposed CBC parameter-based prognostic model and the associated web-application, can help the medical doctors to improve the management by early prediction of mortality risk of the COVID-19 patients in the low-resource countries.

Detection and Severity Classification of COVID-19 in CT Images Using Deep Learning.

Detecting COVID-19 at an early stage is essential to reduce the mortality risk of the patients. In this study, a cascaded system is proposed to segment the lung, detect, localize, and quantify COVID-19 infections from computed tomography images. An extensive set of experiments were performed using Encoder-Decoder Convolutional Neural Networks (ED-CNNs), UNet, and Feature Pyramid Network (FPN), with different backbone (encoder) structures using the variants of DenseNet and ResNet. The conducted experiments for lung region segmentation showed a Dice Similarity Coefficient (DSC) of 97.19% and Intersection over Union (IoU) of 95.10% using U-Net model with the DenseNet 161 encoder. Furthermore, the proposed system achieved an elegant performance for COVID-19 infection segmentation with a DSC of 94.13% and IoU of 91.85% using the FPN with DenseNet201 encoder. The proposed system can reliably localize infections of various shapes and sizes, especially small infection regions, which are rarely considered in recent studies. Moreover, the proposed system achieved high COVID-19 detection performance with 99.64% sensitivity and 98.72% specificity. Finally, the system was able to discriminate between different severity levels of COVID-19 infection over a dataset of 1110 subjects with sensitivity values of 98.3%, 71.2%, 77.8%, and 100% for mild, moderate, severe, and critical, respectively.

SSG-LUGIA: Single Sequence based Genome Level Unsupervised Genomic Island Prediction Algorithm.

BACKGROUND: Genomic Islands (GIs) are clusters of genes that are mobilized through horizontal gene transfer. GIs play a pivotal role in bacterial evolution as a mechanism of diversification and adaptation to different niches. Therefore, identification and characterization of GIs in bacterial genomes is important for understanding bacterial evolution. However, quantifying GIs is inherently difficult, and the existing methods suffer from low prediction accuracy and precision-recall trade-off. Moreover, several of them are supervised in nature, and thus, their applications to newly sequenced genomes are riddled with their dependency on the functional annotation of existing genomes. RESULTS: We present SSG-LUGIA, a completely automated and unsupervised approach for identifying GIs and horizontally transferred genes. SSG-LUGIA is a novel method based on unsupervised anomaly detection technique, accompanied by further refinement using cues from signal processing literature. SSG-LUGIA leverages the atypical compositional biases of the alien genes to localize GIs in prokaryotic genomes. SSG-LUGIA was assessed on a large benchmark dataset `IslandPick' and on a set of 15 well-studied genomes in the literature and followed by a thorough analysis on the well-understood Salmonella typhi CT18 genome. Furthermore, the efficacy of SSG-LUGIA in identifying horizontally transferred genes was evaluated on two additional bacterial genomes, namely, those of Corynebacterium diphtheria NCTC13129 and Pseudomonas aeruginosa LESB58. SSG-LUGIA was examined on draft genomes and was demonstrated to be efficient as an ensemble method. CONCLUSIONS: Our results indicate that SSG-LUGIA achieved superior performance in comparison to frequently used existing methods. Importantly, it yielded a better trade-off between precision and recall than the existing methods. Its nondependency on the functional annotation of genomes makes it suitable for analyzing newly sequenced, yet uncharacterized genomes. Thus, our study is a significant advance in identification of GIs and horizontally transferred genes. SSG-LUGIA is available as an open source software at https://nibtehaz.github.io/SSG-LUGIA/.

An Agent-Based Modeling of COVID-19: Validation, Analysis, and Recommendations.

The coronavirus disease 2019 (COVID-19) has resulted in an ongoing pandemic worldwide. Countries have adopted non-pharmaceutical interventions (NPI) to slow down the spread. This study proposes an agent-based model that simulates the spread of COVID-19 among the inhabitants of a city. The agent-based model can be accommodated for any location by integrating parameters specific to the city. The simulation gives the number of total COVID-19 cases. Considering each person as an agent susceptible to COVID-19, the model causes infected individuals to transmit the disease via various actions performed every hour. The model is validated by comparing the simulation to the real data of Ford County, KS, USA. Different interventions, including contact tracing, are applied on a scaled-down version of New York City, USA, and the parameters that lead to a controlled epidemic are determined. Our experiments suggest that contact tracing via smartphones with more than 60% of the population owning a smartphone combined with city-wide lockdown results in the effective reproduction number (R t ) to fall below 1 within 3 weeks of intervention. For 75% or more smartphone users, new infections are eliminated, and the spread is contained within 3 months of intervention. Contact tracing accompanied with early lockdown can suppress the epidemic growth of COVID-19 completely with sufficient smartphone owners. In places where it is difficult to ensure a high percentage of smartphone ownership, tracing only emergency service providers during a lockdown can go a long way to contain the spread. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at (10.1007/s12559-020-09801-w).

A contour property based approach to segment nuclei in cervical cytology images.

BACKGROUND: Segmentation of nuclei in cervical cytology pap smear images is a crucial stage in automated cervical cancer screening. The task itself is challenging due to the presence of cervical cells with spurious edges, overlapping cells, neutrophils, and artifacts. METHODS: After the initial preprocessing steps of adaptive thresholding, in our approach, the image passes through a convolution filter to filter out some noise. Then, contours from the resultant image are filtered by their distinctive contour properties followed by a nucleus size recovery procedure based on contour average intensity value. RESULTS: We evaluate our method on a public (benchmark) dataset collected from ISBI and also a private real dataset. The results show that our algorithm outperforms other state-of-the-art methods in nucleus segmentation on the ISBI dataset with a precision of 0.978 and recall of 0.933. A promising precision of 0.770 and a formidable recall of 0.886 on the private real dataset indicate that our algorithm can effectively detect and segment nuclei on real cervical cytology images. Tuning various parameters, the precision could be increased to as high as 0.949 with an acceptable decrease of recall to 0.759. Our method also managed an Aggregated Jaccard Index of 0.681 outperforming other state-of-the-art methods on the real dataset. CONCLUSION: We have proposed a contour property-based approach for segmentation of nuclei. Our algorithm has several tunable parameters and is flexible enough to adapt to real practical scenarios and requirements.

MultiResUNet : Rethinking the U-Net architecture for multimodal biomedical image segmentation.

In recent years Deep Learning has brought about a breakthrough in Medical Image Segmentation. In this regard, U-Net has been the most popular architecture in the medical imaging community. Despite outstanding overall performance in segmenting multimodal medical images, through extensive experimentations on some challenging datasets, we demonstrate that the classical U-Net architecture seems to be lacking in certain aspects. Therefore, we propose some modifications to improve upon the already state-of-the-art U-Net model. Following these modifications, we develop a novel architecture, MultiResUNet, as the potential successor to the U-Net architecture. We have tested and compared MultiResUNet with the classical U-Net on a vast repertoire of multimodal medical images. Although only slight improvements in the cases of ideal images are noticed, remarkable gains in performance have been attained for the challenging ones. We have evaluated our model on five different datasets, each with their own unique challenges, and have obtained a relative improvement in performance of 10.15%, 5.07%, 2.63%, 1.41%, and 0.62% respectively. We have also discussed and highlighted some qualitatively superior aspects of MultiResUNet over classical U-Net that are not really reflected in the quantitative measures.