Errors In Judgment Research Articles

Performance Analysis of Classification and Detection of Brain Tumor MRI Images Using Resnet50 Deep Residual U-Net

Background: This research adds to the growing body of work demonstrating the vital role of image categorization in the medical sector. The efficiency of illness diagnosis is being greatly enhanced using image classification. A brain tumor is a collection of abnormal cells in the brain. Diagnostic precision is required when looking for a tumor in the brain because even the smallest error in human judgement can have disastrous results. To get around this problem, the medical community has implemented an automated brain tumor segmentation system. A variety of methods are employed to segment a brain tumor, but the results are inconsistent. To improve the quality of MRI images, we present our findings in this paper. Deep learning methods for image segmentation and classification are discussed in this paper.
 Methods: In this paper we a very robust deep learning method for image segmentation and classification. For image classification, we are employing the enhanced faster R-CNN method. The ResNet50 model is used to differentiate between tumor and non-tumor images. The next step in this framework is to use Deep Residual UNET for segmentation. RESUNET is an FCNN that maximizes efficiency. The proposed method works well in terms of its ability to find and classify things accurately.
 Results: The accuracy rate for identifying tumours in MRI scans using the proposed technique is 94.23%. Using transfer learning with Resnet50 as the base model and the use of discriminative learning rates, our model achieved superior results than other recent models.
 Conclusion: Within the scope of this study, we have integrated the residual networks with the U-Net to make the network stronger. This strategy improves the efficiency of classification and segmentation tools. To achieve a higher level of accuracy, the model may be trained further, or additional data may be applied in the training process.

DR.BENCH: Diagnostic Reasoning Benchmark for Clinical Natural Language Processing

The meaningful use of electronic health records (EHR) continues to progress in the digital era with clinical decision support systems augmented by artificial intelligence. A priority in improving provider experience is to overcome information overload and reduce the cognitive burden so fewer medical errors and cognitive biases are introduced during patient care. One major type of medical error is diagnostic error due to systematic or predictable errors in judgement that rely on heuristics. The potential for clinical natural language processing (cNLP) to model diagnostic reasoning in humans with forward reasoning from data to diagnosis and potentially reduce cognitive burden and medical error has not been investigated. Existing tasks to advance the science in cNLP have largely focused on information extraction and named entity recognition through classification tasks. We introduce a novel suite of tasks coined as Diagnostic Reasoning Benchmarks, Dr.Bench, as a new benchmark for developing and evaluating cNLP models with clinical diagnostic reasoning ability. The suite includes six tasks from ten publicly available datasets addressing clinical text understanding, medical knowledge reasoning, and diagnosis generation. DR.BENCH is the first clinical suite of tasks designed to be a natural language generation framework to evaluate pre-trained language models for diagnostic reasoning. The goal of DR. BENCH is to advance the science in cNLP to support downstream applications in computerized diagnostic decision support and improve the efficiency and accuracy of healthcare providers during patient care. We fine-tune and evaluate the state-of-the-art generative models on DR.BENCH. Experiments show that with domain adaptation pre-training on medical knowledge, the model demonstrated opportunities for improvement when evaluated in DR. BENCH. We share DR. BENCH as a publicly available GitLab repository with a systematic approach to load and evaluate models for the cNLP community. We also discuss the carbon footprint produced during the experiments and encourage future work on DR.BENCH to report the carbon footprint.

Complications and treatment errors in implant positioning in the aesthetic zone: Diagnosis and possible solutions.

Incorrect implant positioning can lead to functional and aesthetic compromise. Implant positioning errors can occur in three dimensions: mesiodistal, corono-apical, and orofacial. Treatment solutions to manage adverse outcomes through positioning errors require an understanding of the underlying conditions and of those factors that may have led to the error being committed in the first place. These types of complications usually occur because of human factors. If errors do occur with adverse aesthetic outcomes, they are difficult and sometimes impossible to correct. Connective tissue grafts to reverse recession defects are only feasible in defined situations. The option to remove and replace the implant may be the only recourse, provided the removal process does not further compromise the site. Error in judgment by the clinician.

Sizing up the crowd: Assessing spatial integration difficulties in body size judgements across eating disorder symptomatology.

Body size judgements are frequently biased, or inaccurate, and these errors are further exaggerated for individuals with eating disorders. Within the eating disorder literature, it has been suggested that exaggerated errors in body size judgements are due to difficulties with integration. Across two experiments, we developed a novel integration task, named the Ebbinghaus Illusion for Bodies in Virtual Reality (VR), to assess whether nearby bodies influence the perceived size of a single body. VR was used to simulate the appearance of a small crowd around a central target body. In Experiment 1 (N = 412), participants were required to judge the size of a central female target within a crowd. Experiment 1 revealed an Ebbinghaus Illusion, in which a central female appeared larger when surrounded by small distractors, but comparatively smaller when surrounded by large distractors. In other words, the findings of Experiment 1 demonstrate that surrounding crowd information is integrated when judging an individual's body size; a novel measure of spatial integration (i.e., an Ebbinghaus Illusion for Bodies in VR). In Experiment 2 (N = 96), female participants were selected based on high (n = 43) and low (n = 53) eating disorder symptomatology. We examined whether the magnitude of this illusion would differ amongst those with elevated versus low eating disorder symptomatology, in accordance with weak central coherence theory, with the high symptomatology group displaying less spatial integration relative to the low group. The results of Experiment 2 similarly found an Ebbinghaus Illusion for Bodies in VR. However, illusion magnitude did not vary across high and low symptomatology groups. Overall, these findings demonstrate that surrounding crowd information is integrated when judging individual body size; however, those with elevated eating disorder symptomatology did not show any integration deficit on this broader measure of spatial integration.

LESÕES ESPORTIVAS DE ESPORTES NACIONAIS TRADICIONAIS EM ESTUDANTES UNIVERSITÁRIOS

ABSTRACT Introduction: There are 56 ethnic groups in China. Many ethnic groups have their sports, which can cause sports injuries to the human body. Objective: This paper analyzes the impact of national sports on college students' physical health, investigates the most common sports injuries in college students, and specifies scientifically-based prevention and intervention measures. Methods: 312 college students who regularly participate in various national sporting events at 10 colleges and universities volunteered to investigate common injuries sustained by participating in national sporting activities. The main injury factors were analyzed using statistical methods. Results: The investigation shows that college students' errors of judgment regarding accidents, risky behaviors, and poor sports facilities are the main causes of sports injuries. Injuries are mainly concentrated in the lower limbs and pelvic girdle. Conclusion: College students' participation in national sports activities is conducive to improving their physical and mental health. However, attention to preventing injury factors in the sports process is needed to ensure safe and healthy participation in national sports. Level of evidence II; Therapeutic studies - investigation of treatment outcomes.

You get what you screen for: Standards for experimental design and data fitting in drug discovery.

A common mantra in drug discovery is that "You get what you screen for." This is not a promise that you will always get an effective drug candidate, but rather a warning that inaccuracies in your protocol for screening will more likely produce a compound that fails to be an effective candidate because it matches the properties of your screen, not the desired features of an ideal lead compound. It is with this in mind that we examine the current protocols for evaluating drug candidates and highlight some deficiencies while pointing the way to better methods. Many of the errors in data fitting can be rectified by abandoning the traditional equation-based data fitting methods and adopting the more rigorous mechanism-based fitting afforded by computer simulation based on numerical integration of rate equations. Using these methods bypasses the errors in judgement in choosing the appropriate equation for data fitting and the approximations required to derive those equations. In this chapter we outline the limitations and systematic errors in conventional methods of data fitting and illustrate the advantages of computer simulation and introduce the methods of analysis.

Gastrovigilance: A Close Watch on Gastrointestinal and Hepatic Disorders- An Indian Perspective

Gastrointestinal and hepatic disorders account for about 25% of consultations among general practitioners in India. Errors in clinical judgement and hesitancy in recommending necessary tests owing to lack of health insurance could result in delayed diagnosis and increased patient morbidity and mortality. Clinicians should thus be well equipped with effective strategies for skilful diagnosis and in a position to weigh the benefit-risk-ratio of recommending pertinent and disregarding less useful diagnostic tests. 'Gastrovigilance' includes disease-specific training for recognising risk factors, algorithms and referral pathways. This narrative review focuses on the common challenges or errors in managing these conditions in Indian clinical practice and their proposed solutions. Literature searches were performed using PubMed/MEDLINE and Google Scholar following the shortlisted gastrointestinal conditions. Based on the published literature and expertise of the senior gastroenterologists, improving disease-specific knowledge can enhance rates of correct diagnosis. Improved screening and patient education can reduce the risk of presentation at advanced stages and consequently improve prognosis. Another significant contributory factor is the patient-physician interaction which affects every stage of the disease management and methods to improve it, therefore vital in improving gastrointestinal and hepatic disease conditions. The most important means of improving gastrovigilance is optimising knowledge access in primary care. This shall improve diagnostic accuracy and reduce the burden of misdiagnosis. In the current narrative review, we have tried to elucidate the concept of gastrovigilance for gastrointestinal and hepatic conditions and substantiate it with published evidence.

Avoiding cognitive bias : example of quantified assessment of muscle strength

Optimizing treatments requires the prevention of diagnostic errors. The use of shortcuts, cognitive biases, may lead to errors of judgement that can impair clinical reasoning and distort decision-making. Objective, quantifiable and comparable assessment is a safeguard against this, and for instance force quantification is used for this purpose. We will present here the use of hand-held dynamometers with recommendations for best practice as well as simple but essential tools for interpreting the results. Indeed, the use of these easy-to-use dynamometers requires rigour to ensure the quality of measurements and data analysis by/for the clinician. The use of this equipment should be widely promoted and democratized in informed clinical practice.

Evaluation of spoken English self-study system considering speech knowledge recognition algorithm

Abstract This paper builds and evaluates a self-learning system for spoken English pronunciation suitable for PC mobile terminal or mobile terminal users, and the core technology of speech recognition is built on Hidden Markov Model (HMM), which is used to decode the speech signal in spoken English learning. This paper studies the related speech recognition theory and signal processing technology, builds a comprehensive English self-learning system in more complex situations and more user types and conducts a comprehensive evaluation of the system. The results show that the overall accuracy of the HMM model in the spoken English recognition and evaluation system built in this paper is good, and the accuracy of the input audio for people of all ages is greater than 90%. In the younger population, the accuracy of male speech signals was the highest in both closed space and open space, reaching 98.12% and 96.53%, respectively. In addition, it is observed that the accuracy gradually decreases as more wrong judgements are made on the speech input signal. When the evaluation is poor, the accuracy of the scoring results decreases to 55%, and when there are fewer voice judgement errors, it is in the excellent range, where the accuracy reached 88%.

What does a "good life" mean for people living with dementia? A protocol for a think-aloud study informing the value of care.

Economic evaluation currently focuses almost exclusively on the maximization of health, using the Quality-Adjusted Life-Year (QALY) framework with instruments such as the EQ-5D, with a limited number of health-focused dimensions providing the assessment of health benefit. This evaluative framework is likely to be insufficient for setting priorities in dementia care because of its exclusive concern with health. Data are also often collected from the perspective of a proxy, limiting the voice of those living with dementia in decision-making. This protocol describes a research project that aims to gather the perspectives of people living with dementia, their insights, and preferences for assessing their quality of life to inform economic evaluation outcome measurement and design with a goal of creating a more robust evidence base for the value of healthcare services. Specifically, this study will elucidate what a "good life" means to people living with dementia and how well instruments currently used in economic evaluation meet this description. This project will further test the acceptability of capability wellbeing instruments as self-report instruments and compare them to generic and dementia-specific preference-based instruments. People living with dementia, diagnosed, or waiting to receive a formal diagnosis and with the capacity to participate in research, will be invited to participate in an hour "think aloud" interview. Participants will be purposefully selected to cover a range of dementia diagnoses, age, and sex, recruited through the integrated care, geriatric, and post-diagnostic clinics at St James' and Tallaght University Hospitals and dementia support groups in the Ireland. During the interview, participants will be invited to reflect on a "good life" and "think aloud" while completing four economic quality of life instruments with a perspective that goes beyond health (AD-5D/QOL-AD, AQOL-4D, ICECAP-O, ICECAP-SCM). An interviewer will then probe areas of difficulty when completing the instruments in a semi-structured way. The analysis will identify the frequency of errors in comprehension, retrieval, judgment, and response from verbatim transcripts. Qualitative data will be analyzed using constant comparison. The St James's Hospital and Tallaght University Hospital Joint Research Ethics Committee approved the study (Approval Date: 11 April 2022).

A Survey on Tools and Techniques for Localizing Abnormalities in X-ray Images Using Deep Learning

Deep learning is expanding and continues to evolve its capabilities toward more accuracy, speed, and cost-effectiveness. The core ingredients for getting its promising results are appropriate data, sufficient computational resources, and best use of a particular algorithm. The application of these algorithms in medical image analysis tasks has achieved outstanding results compared to classical machine learning approaches. Localizing the area-of-interest is a challenging task that has vital importance in computer aided diagnosis. Generally, radiologists interpret the radiographs based on their knowledge and experience. However, sometimes, they can overlook or misinterpret the findings due to various reasons, e.g., workload or judgmental error. This leads to the need for specialized AI tools that assist radiologists in highlighting abnormalities if exist. To develop a deep learning driven localizer, certain alternatives are available within architectures, datasets, performance metrics, and approaches. Informed decision for selection within the given alternative can lead to batter outcome within lesser resources. This paper lists the required components along-with explainable AI for developing an abnormality localizer for X-ray images in detail. Moreover, strong-supervised vs weak-supervised approaches have been majorly discussed in the light of limited annotated data availability. Likewise, other correlated challenges have been presented along-with recommendations based on a relevant literature review and similar studies. This review is helpful in streamlining the development of an AI based localizer for X-ray images while extendable for other radiological reports.

Customizable Health Policy System

In order to account for the role that hospitals play in health policy, this research paper will use and build customized health policy theory. The way in which health services are provided is heavily influenced by health policies. Due to its inflexible structure and misguiding policies, it is a well-known issue with the purchase of health insurance. Due to this, the appropriate resources are not allocated at the appropriate time, which results in reimbursement failure, health burden, and financial losses as a result of error in judgement or misunderstanding. Therefore, personalizing health policies to meet the requirements of specific people or groups is possible. The framework for making decisions to improve people's health status is known as health policy. The significance and problems with developing health policies are discussed in detail in this paper, along with methodological concerns. Methodological concerns are cited, and the majority of the works in this field are cited. After introducing the subject and outlining the need for health policy, it is explained from several academic perspectives and what a health policy is. The stages of the cycle for developing policies are discussed. The health policy development cycle is described in this review paper, along with the methodological concerns it raises. The study places further emphasis on evidence-based health policy and then describes the lessons that may be drawn from prior attempts at developing policy. The report finishes by outlining the difficulties in creating and implementing a successful health strategy.

Reframing Humanist Tragedy in The Tiniest Thing

Tragedy is defined by Aristotle as a self-fulfilling prophecy, inciting fear and pity in an audience through a hero’s error of judgement, or hamartia. Anthropocentric climate change may likewise be viewed in similar terms, born out of the limitations of the humanist paradigm. Yet in an age of climate catastrophe, how might theatre represent this reality without reinforcing the same humanist logic of privileging human suffering? As a playwright, I have long grappled with how best to dramatise climate change: a phenomenon that seems beyond the scope of human-centred drama. At the same time, the Anthropocene is by definition a human-created problem, and the emotional impact of our doom-laden future bears a tangible human effect. When choosing a form, then, for my own new play about this topic, something of a balance seemed important to me: a human-centred approach that might nonetheless recontextualise human suffering within a more earthly timescale. My resulting new play, The Tiniest Thing, is a middle-class Australian family drama that is rudely interrupted by the natural world. As a forest emerges from a pantry, long grass appears beneath the living room carpet, and dead birds begin to fall from the ceiling, the human characters refuse to see what is right in front of their eyes – until one character, Susan, begins to let the outside world in. Ultimately concerned with the politics of perception, The Tiniest Thing asks: is it possible for humans to perceive an objective reality, or do we always choose what we want to believe? And how might rigid ideologies become our own hamartia in the face of climate catastrophe? I view these questions within the context of bringing an eco-critical dramaturgy to my playwriting, primarily through my use of structure, contrasting deep ‘planetary’ time with the ‘human’ time of the unfolding plot, inspired by the Arctic Cycle plays of Chantal Bilodeau. By seeking to show two different realities at once, I hope to evoke a world on stage in which the same phenomena are perceived by the human characters in vastly different ways, reframing their suffering within a wider ecological context.

Improving Workplace Judgments by Reducing Noise: Lessons Learned from a Century of Selection Research

Some assert that noise (i.e., unwanted variance) is the most neglected yet most important source of error in judgment. We suggest that this problem was discovered nearly 100 years ago in the area of personnel selection and that a century of selection research has shown that noise can be demonstrably reduced by structuring the process (i.e., decomposing the component parts, agreeing on standards, and applying those standards consistently) and by aggregating judgments independently. Algorithms can aid significantly in this process but are often confused with methods that, in their current form, can substantially increase noise in judgment (e.g., artificial intelligence and machine learning).

Hysteroscopic dilation techniques in hysteroscopic adhesiolysis.

Intrauterine adhesion (IUA) is mainly caused by intrauterine operations such as pregnancy-related curettage and hysteroscopic surgery, resulting in the trauma to the basal layer of the endometrium. Hysteroscopic adhesiolysis is a crucial step in the comprehensive treatment of IUA, and the most common complication is uterine perforation. More than half of all uterine perforations occur during the hysteroscopy or probe/dilator pass through the internal os. Furthermore, inappropriate surgical procedures may lead to endometrial injury, recurrence or even aggravation of adhesions, and complications such as cervix laceration and false passage formation. This study aims to explore the usage of the hysteroscopic dilatation techniques to dilate the internal os and lower uterine segment, which is via hysteroscopy entering the internal os laterally and swinging, or by directly opening the forceps or scissors and bluntly spreading dissection under direct hysteroscopic vision. By using the hysteroscopic dilatation techniques, we intend to improve the effectiveness and safety of cervical dilation in patients with IUA in the internal os and/or lower uterine segment. A total of 282 patients with adhesions in the internal os or lower uterine segment underwent HA in the Third Xiangya Hospital of Central South University from January 2020 to June 2021 were included, ranging from 21 to 46 (33.0±4.8) years old in age and 5 to 12 in the American Fertility Society score. Among them, there were 2 cases of false passage formation caused by traditional dilatation in other hospitals. All patients underwent hysteroscopy with integrated hysteroscopy with 5Fr instrument channel and 4.9 mm outer sheath diameter. The internal orifice of cervix and the lower segment of uterine cavity were dilated under the microscope. After the hysteroscopy entered the uterine cavity, the separation of uterine cavity adhesion and the placement of uterine contraceptive ring or uterine stent into the uterine cavity were performed routinely. Age, surgical records, and surgical videos of all included cases were collected. The success rate of dilation and the incidence of surgical complications were assessed. In all cases, the hysteroscopys successfully entered into the uterine cavity by using the hysteroscopic dilatation techniques without failure and switching to cervical dilators. In the 2 cases of false passage due to previous cervical dilation, the uterine cavity was identified and found successfully under direct hysteroscopic vision. During the whole surgery, the vision was clear, and no complications (such as cervix laceration, false passage formation, uterine perforation or water intoxication) occurred. One to 3 months postoperative hysteroscopy revealed no significant fibrotic stenosis in the internal os and lower uterine segment. The hysteroscopic dilation techniques are a strategy for separation methods that is following structural hierarchy anatomy in the mode of "see and treat" for the adhesion in the internal os and uterine cavity under direct hysteroscopic vision. This method not only has ultrasound guidance, but also has the judgment of structural hierarchy anatomy under direct hysteroscopic vision, so there is less chance of anatomical level judgment error. This method makes full use of the hysteroscopic judgement of the experienced hysteroscopic surgeons, so that surgeons can timely find and avoid re-entering the old false passage caused by previous surgery. The adhesions in the internal os and lower uterine segment were separated by the hysteroscopic dilation techniques. In this way, the damage to the endometrium caused by forced insertion of the hysteroscopy can be avoided. Meticulous separation of adhesions and cervical dilation under direct hysteroscopic vision can effectively reduce the occurrence of surgical complications such as false passage formation, cervical laceration, and uterine perforation. The use of mini-hysteroscopy eliminates the need for preoperative cervical preparation, avoiding associated risks and side effects. Moreover, for patients with adhesions in the internal os and lower uterine segment, preoperative cervical preparation is not effective in cervical dilation, while the hysteroscopic dilation techniques are effective, with higher patient acceptance due to the absence of preoperative cervical preparation. For the skilled hysteroscopic surgeons, the hysteroscopic dilation technique is easy to operate and worthy of clinical application.

Clinical, Imaging, and Prognostic Differences of Neurosyphilis in HIV and Non-HIV Individuals

Background Neurosyphilis remains an important health issue, particularly in populations at risk like men who have sex with men. Our aim was determining differences in the clinical presentation and neuroimaging findings in HIV-positive and non-HIV neurosyphilis patients. Methods This is a retrospective study that included 30 patients who were classified either as non-HIV (n = 16) or HIV positive (n = 14) during the period 1997–2017 at the Instituto Nacional de Neurología y Neurocirugía. Results The main clinical signs in non-HIV patients were dementia (56%, P = 0.02), dementia plus neuro-ophthalmic alterations (50%, P = 0.01), delirious ideas (62.5%, P = 0.03), and judgment errors (44%, P = 0.02). In contrast, neuro-ophthalmic alterations were predominant in HIV-positive patients (85.7%, P = 0.02) along with pachymeningitis (57.1%, P = 0.021) and optic neuritis (57.1%, P = 0.023), as determined by neuroimaging. No significant differences in the cytochemical characteristics of cerebrospinal fluid were observed. Conclusions Dementia was predominant in non-HIV subjects, whereas neuro-ophthalmic lesions were observed earlier and more frequently in HIV-positive patients. Our results highlight the importance of visual signs in HIV-positive patients as an early manifestation of neurosyphilis, and the relevant role of the cerebrospinal fluid Venereal Disease Research Laboratory test in the evaluation of patients with dementia.

A hybrid framework for modelling domains using quantitative covariates

Domains define the boundaries of mineralisation zones, within which the grade distribution of the target minerals can be quantified via an established mineral resource estimation procedure. Available domain modelling techniques include manual interpretation, implicit modelling and advanced geostatistical approaches. In mining applications, the most commonly used method is manual domaining, which is labour-intensive and prone to subjective judgement errors. In addition, the output is largely deterministic and ignores the significant uncertainty associated with the domain interpretation and boundary definitions. There is, therefore, a need for a more objective framework that can automatically define mineral domains and quantify the associated uncertainty. This paper describes such a framework, which consists of a hybrid approach based on simulated grade distributions and a machine learning (ML) classification technique, XGBoost, trained on lithological properties. Data from an Iron Oxide Copper Gold (IOCG) deposit are used as a case study to demonstrate the application of the proposed method. The study shows that the approach can handle complex multi-class problems with imbalanced features, and it can quantify the uncertainty of domain boundaries. A noise filtering method is used as a pre-processing step to improve the performance of the ML classification, especially in the case of problematic classes where domain boundaries are difficult to predict due to the similarity in geological characteristics.

The changes in subjective visual vertical after otolith reduction in patients with BPPV

Objective:To observe and analyze the changes in subjective visual vertical（SVV） after otolith reduction in patients with BPPV. Methods:46 patients with confirmed BPPV recieving successful otolith reduction were selected as the test group. 31 cases of posterior canal stones and 15 cases of horizontal semicircular canal stones, 29 cases of right ear and 17 cases of left ear. Fifty cases of healthy young volunteers were in the control group. Using the virtual reality SVV examination system, 0° SVV in the positive head were tested in the test group patients before and after the reduction of SVV , and were tested in the control group .The deviation angle of the SVV before and after the otolith reduction in the test group were analyzed. Results:Before otoliths reduction, the SVV was （0.08±3.83）° of right BPPV and was （-1.69±2.23）° of left BPPV. After otoliths reduction, the SVV was （-1.52±3.74）° of right BPPV and was （-1.04±2.50）° of left BPPV. In the control group, the SVV was（-1.57±2.28）° . The changes of SVV deflection angle between the control group and the right BPPV before the otolith reduction, and before and after the otolith reduction in the right BPPV were analyzed, and the differences were all statistically significant. There was no significant difference in SVV deflection angle between the left BPPV（before and after reduction） and the control group. In the test group, after the otolith reduction, 18 cases had larger bias angles, 28 cases had smaller bias angle among which 13 cases the deviation angle even turned to the contralateral side. Conclusion:Utriculare dysfunction in patients with BPPV leads to the judgment error of SVV. Reduction of otolithoid can cause new stimulation to the eutricule and affect its functional status. SVV detection can provide help for the evaluation of utricular function in patients with BPPV.

A Soft Label Deep Learning to Assist Breast Cancer Target Therapy and Thyroid Cancer Diagnosis.

According to the World Health Organization Report 2022, cancer is the most common cause of death contributing to nearly one out of six deaths worldwide. Early cancer diagnosis and prognosis have become essential in reducing the mortality rate. On the other hand, cancer detection is a challenging task in cancer pathology. Trained pathologists can detect cancer, but their decisions are subjective to high intra- and inter-observer variability, which can lead to poor patient care owing to false-positive and false-negative results. In this study, we present a soft label fully convolutional network (SL-FCN) to assist in breast cancer target therapy and thyroid cancer diagnosis, using four datasets. To aid in breast cancer target therapy, the proposed method automatically segments human epidermal growth factor receptor 2 (HER2) amplification in fluorescence in situ hybridization (FISH) and dual in situ hybridization (DISH) images. To help in thyroid cancer diagnosis, the proposed method automatically segments papillary thyroid carcinoma (PTC) on Papanicolaou-stained fine needle aspiration and thin prep whole slide images (WSIs). In the evaluation of segmentation of HER2 amplification in FISH and DISH images, we compare the proposed method with thirteen deep learning approaches, including U-Net, U-Net with InceptionV5, Ensemble of U-Net with Inception-v4, Inception-Resnet-v2 encoder, and ResNet-34 encoder, SegNet, FCN, modified FCN, YOLOv5, CPN, SOLOv2, BCNet, and DeepLabv3+ with three different backbones, including MobileNet, ResNet, and Xception, on three clinical datasets, including two DISH datasets on two different magnification levels and a FISH dataset. The result on DISH breast dataset 1 shows that the proposed method achieves high accuracy of 87.77 ± 14.97%, recall of 91.20 ± 7.72%, and F1-score of 81.67 ± 17.76%, while, on DISH breast dataset 2, the proposed method achieves high accuracy of 94.64 ± 2.23%, recall of 83.78 ± 6.42%, and F1-score of 85.14 ± 6.61% and, on the FISH breast dataset, the proposed method achieves high accuracy of 93.54 ± 5.24%, recall of 83.52 ± 13.15%, and F1-score of 86.98 ± 9.85%, respectively. Furthermore, the proposed method outperforms most of the benchmark approaches by a significant margin (p <0.001). In evaluation of segmentation of PTC on Papanicolaou-stained WSIs, the proposed method is compared with three deep learning methods, including Modified FCN, U-Net, and SegNet. The experimental result demonstrates that the proposed method achieves high accuracy of 99.99 ± 0.01%, precision of 92.02 ± 16.6%, recall of 90.90 ± 14.25%, and F1-score of 89.82 ± 14.92% and significantly outperforms the baseline methods, including U-Net and FCN (p <0.001). With the high degree of accuracy, precision, and recall, the results show that the proposed method could be used in assisting breast cancer target therapy and thyroid cancer diagnosis with faster evaluation and minimizing human judgment errors.

Investigation of contaminant profile in highway stormwater runoff and risk assessment by statistical analysis

Highway stormwater runoff pollution has become a severe risk factor for water bodies nowadays. The conventional risk analysis protocols for directly discharging highway runoff are prone to systematic and judgmental errors. Therefore, a numeric and straightforward risk assessment protocol has been developed in this study that minimizes the errors. For this study, three highway segments were selected in the city of Rawalpindi, Pakistan. Event mean concentrations have been used as baseline numeric values for calculating the risk of discharging highway stormwater directly into the water bodies. These values are also correlated with highway characteristics (area, slope, and traffic count) and storm characteristics (storm depth, cumulative runoff volume, antecedent dry days, and cumulative flow). The highway stormwater was monitored for organics, metals, solids, and macro-nutrients at three highway sections. The event mean concentrations of dissolved organic carbon (50–145 mg/L), total suspended solids (1500–3900 mg/L), chromium (0.25–0.45 mg/L), and lead (0.1–0.8 mg/L) are found to be higher than the environmental quality standards. The risk assessment was conducted by the analysis of variance. The analysis showed that the highway characteristics significantly affect contaminant concentrations, but storm characteristics on contaminant concentrations are not found to be significant. Total suspended solids are the most threatening contaminant in highway runoffs. The study concluded that the risk from contaminants in highway stormwater depends particularly on the specific highway sections’ properties. The first flush portion (initial 25% of runoff) of highway runoff poses a higher threat to the receiving environment than the later runoff volumes.