RESUMO
BACKGROUND AND AIM: Inpatients undergoing colonoscopy experience a higher-than-average rate of inadequate bowel preparation (compared to outpatients) leading to canceled procedures, increased stress on the patient, increased time in hospital, and increased cost to the healthcare system. The aim of this scoping review was to identify research surrounding inpatient bowel preparation and to identify modifiable and non-modifiable factors that influence the adequacy of bowel preparation in hospitalized patients undergoing colonoscopy and establish areas where nursing interventions may help improve overall bowel preparation rates. METHODS: An initial search of MEDLINE, CINAHL, Scopus, and Embase was undertaken to identify seed articles, followed by a structured search using keywords and subject headings. Studies conducted between 2000 and 2022 and published in English were included. A total of 37 full-text studies were screened for inclusion, with 22 meeting inclusion criteria. RESULTS: Advanced age, decreased mobility, constipation, extended length of stay, and multiple comorbidities were identified as non-modifiable factors associated with inadequate bowel preparation. Narcotic use, failure to follow preparation instruction, and delayed time to colonoscopy were identified as modifiable factors associated with poor bowel preparation. CONCLUSIONS: Educational interventions and interprofessional programs, using a multifaceted approach, increase the odds of adequate bowel preparation, including nursing tip sheets, troubleshooting flowsheets, and bowel movement assessment scoring.
RESUMO
This study is aimed at examining the impact of ChatGPT on pediatric endocrine and metabolic conditions, particularly in the areas of screening and diagnosis, in both Chinese and English modes. A 40-question questionnaire covering the four most common pediatric endocrine and metabolic conditions was posed to ChatGPT in both Chinese and English three times each. Six pediatric endocrinologists evaluated the responses. ChatGPT performed better when responding to questions in English, with an unreliable rate of 7.5% compared to 27.5% for Chinese questions, indicating a more consistent response pattern in English. Among the reliable questions, the answers were more comprehensive and satisfactory in the English mode. We also found disparities in ChatGPT's performance when interacting with different target groups and diseases, with improved performance for questions posed by clinicians in English and better performance for questions related to diabetes and overweight/obesity in Chinese for both clinicians and patients. Language comprehension, providing incomprehensive answers, and errors in key data were the main contributors to the low scores, according to reviewer feedback. CONCLUSION: Despite these limitations, as ChatGPT continues to evolve and expand its network, it has significant potential as a practical and effective tool for clinical diagnosis and treatment. WHAT IS KNOWN: ⢠The deep learning-based large-language model ChatGPT holds great promise for improving clinical practice for both physicians and patients and has the potential to increase the speed and accuracy of disease screening and diagnosis, as well as enhance the overall efficiency of the medical process. However, the reliability and appropriateness of AI model responses in specific field remains unclear. ⢠This study focused on the reliability and appropriateness of AI model responses to straightforward and fundamental questions related to the four most prevalent pediatric endocrine and metabolic disorders, for both healthcare providers and patients, in different language scenarios. WHAT IS NEW: ⢠The AI model performed better when responding to questions in English, with more consistent, as well as more comprehensive and satisfactory responses. In addition, we also found disparities in ChatGPT's performance when interacting with different target groups and different diseases. ⢠Despite these limitations, as ChatGPT continues to evolve and expand its network, it has significant potential as a practical and effective tool for clinical diagnosis and treatment.
Assuntos
Inteligência Artificial , Doenças do Sistema Endócrino , Humanos , Doenças do Sistema Endócrino/diagnóstico , Criança , Inquéritos e Questionários , Idioma , Programas de Rastreamento/métodos , Feminino , Pediatria/métodos , Masculino , China/epidemiologiaRESUMO
OBJECTIVES: To investigate the disease spectrum and pathogenic genes of inherited metabolic disorder (IMD) among neonates in Gansu Province of China. METHODS: A retrospective analysis was conducted on the tandem mass spectrometry data of 286 682 neonates who received IMD screening in Gansu Provincial Maternal and Child Health Hospital from January 2018 to December 2021. A genetic analysis was conducted on the neonates with positive results in tandem mass spectrometry during primary screening and reexamination. RESULTS: A total of 23 types of IMD caused by 28 pathogenic genes were found in the 286 682 neonates, and the overall prevalence rate of IMD was 0.63 (1/1 593), among which phenylketonuria showed the highest prevalence rate of 0.32 (1/3 083), followed by methylmalonic acidemia (0.11, 1/8 959) and tetrahydrobiopterin deficiency (0.06, 1/15 927). In this study, 166 variants were identified in the 28 pathogenic genes, with 13 novel variants found in 9 genes. According to American College of Medical Genetics and Genomics guidelines, 5 novel variants were classified as pathogenic variants, 7 were classified as likely pathogenic variants, and 1 was classified as the variant of uncertain significance. CONCLUSIONS: This study enriches the database of pathogenic gene variants for IMD and provides basic data for establishing an accurate screening and diagnosis system for IMD in this region.
Assuntos
Erros Inatos do Metabolismo dos Aminoácidos , Doenças Metabólicas , Criança , Recém-Nascido , Humanos , Estudos Retrospectivos , Doenças Metabólicas/genética , Erros Inatos do Metabolismo dos Aminoácidos/genética , China , Saúde da CriançaRESUMO
Background: Gastroesophageal reflux disease (GERD) is heterogeneous with a varied symptom spectrum and reflux profiles. Its definite diagnosis often requires invasive tools including endoscopy or reflux monitoring. The aim of this study was to investigate the clinical relevance of salivary pepsin detection as a non-invasive screening tool to diagnose GERD of different subtypes. Methods: A total of 77 patients with suspected GERD symptoms and 12 asymptomatic controls were analysed. All participants performed symptom evaluation, upper endoscopy, esophageal manometry, and 24-hour multichannel intraluminal impedance-dual pH probe monitoring. Saliva was self-collected across three different time points: at early fasting, postprandially, and at symptom occurrence. Salivary pepsin levels were measured via Peptest. The optimal threshold of salivary pepsin for diagnosing distal or proximal reflux was determined according to a receiver-operating characteristic curve. Results: The average salivary pepsin concentration of suspected GERD patients was significantly higher than that of controls (100.63 [68.46, 141.38] vs 67.90 [31.60, 115.06] ng/mL, P = 0.044), although no difference was found among patients with different symptom spectrums. The distal reflux group had a higher average pepsin concentration than non-reflux patients (170.54 [106.31, 262.76] vs 91.13 [63.35, 127.63] ng/mL, P = 0.043), while no difference was observed between the distal reflux group and the proximal reflux group. The optimal cut-off value of salivary pepsin concentration for diagnosing pathological distal reflux was 157.10 ng/mL, which was higher than that for diagnosing pathological proximal reflux (122.65 ng/mL). The salivary pepsin concentration was significantly correlated with distal and proximal reflux parameters. Conclusions: Salivary pepsin measurement can help in identifying true GERD with pathological distal reflux or proximal reflux, regardless of different symptom spectrums. A higher threshold should be applied for diagnosing distal reflux than for proximal reflux.
RESUMO
Glaucoma is a chronic eye disease that may lead to permanent vision loss if it is not diagnosed and treated at an early stage. The disease originates from an irregular behavior in the drainage flow of the eye that eventually leads to an increase in intraocular pressure, which in the severe stage of the disease deteriorates the optic nerve head and leads to vision loss. Medical follow-ups to observe the retinal area are needed periodically by ophthalmologists, who require an extensive degree of skill and experience to interpret the results appropriately. To improve on this issue, algorithms based on deep learning techniques have been designed to screen and diagnose glaucoma based on retinal fundus image input and to analyze images of the optic nerve and retinal structures. Therefore, the objective of this paper is to provide a systematic analysis of 52 state-of-the-art relevant studies on the screening and diagnosis of glaucoma, which include a particular dataset used in the development of the algorithms, performance metrics, and modalities employed in each article. Furthermore, this review analyzes and evaluates the used methods and compares their strengths and weaknesses in an organized manner. It also explored a wide range of diagnostic procedures, such as image pre-processing, localization, classification, and segmentation. In conclusion, automated glaucoma diagnosis has shown considerable promise when deep learning algorithms are applied. Such algorithms could increase the accuracy and efficiency of glaucoma diagnosis in a better and faster manner.
RESUMO
INTRODUCTION: Lung cancer is a leading cause of death in patients with cancer. Early diagnosis is crucial to improve the prognosis of patients with lung cancer. Plasma circulating cell-free DNA (cfDNA) contains comprehensive genetic and epigenetic information from tissues throughout the body, suggesting that early detection of lung cancer can be done non-invasively, conveniently, and cost-effectively using high-sensitivity techniques such as sequencing. AREAS COVERED: In this review, we summarize the latest technological innovations, coupled with next-generation sequencing (NGS), regarding genomic alterations, methylation, and fragmentomic features of cfDNA for the early detection of lung cancer, as well as their clinical advances. Additionally, we discuss the suitability of study designs for diagnostic accuracy evaluation for different target populations and clinical questions. EXPERT OPINION: Currently, cfDNA-based early screening and diagnosis of lung cancer faces many challenges, such as unsatisfactory performance, lack of quality control standards, and poor repeatability. However, the progress of several large prospective studies employing epigenetic features has shown promising predictive performance, which has inspired cfDNA sequencing for future clinical applications. Furthermore, the development of multi-omics markers for lung cancer, including genome-wide methylation and fragmentomics, is expected to play an increasingly important role in the future.
Assuntos
Ácidos Nucleicos Livres , Neoplasias Pulmonares , Humanos , Estudos Prospectivos , DNA de Neoplasias/genética , Detecção Precoce de Câncer/métodos , Neoplasias Pulmonares/diagnóstico , Neoplasias Pulmonares/genética , Ácidos Nucleicos Livres/genética , Análise de Sequência de DNA , Biomarcadores Tumorais/genéticaRESUMO
BACKGROUND AND AIM: This study aimed to assess utilization of health-care services in people with decompensated cirrhosis (DC) or hepatocellular carcinoma (HCC) and a "late diagnosis" of hepatitis B or hepatitis C. METHODS: Hepatitis B and C cases during 1997-2016 in Victoria, Australia, were linked with hospitalizations, deaths, liver cancer diagnoses, and medical services. A late diagnosis was defined as hepatitis B or hepatitis C notification occurring after, at the same time, or within 2 years preceding an HCC/DC diagnosis. Services provided during the 10-year period before HCC/DC diagnosis were assessed, including general practitioner (GP) or specialist visits, emergency department presentations, hospital admissions, and blood tests. RESULTS: Of the 25 766 notified cases of hepatitis B, 751 (2.9%) were diagnosed with HCC/DC, and hepatitis B was diagnosed late in 385 (51.3%). Of 44 317 cases of hepatitis C, 2576 (5.8%) were diagnosed with HCC/DC, and hepatitis C was diagnosed late in 857 (33.3%). Although late diagnosis dropped over time, missed opportunities for timely diagnosis were observed. Most people diagnosed late had visited a GP (97.4% for hepatitis B, 98.9% for hepatitis C) or had a blood test (90.9% for hepatitis B, 88.6% for hepatitis C) during the 10 years before HCC/DC diagnosis. The median number of GP visits was 24 and 32, and blood tests 7 and 8, for hepatitis B and C, respectively. CONCLUSIONS: Late diagnosis of viral hepatitis remains a concern, with the majority having frequent health-care service provision in the preceding period, indicating missed opportunities for diagnosis.
Assuntos
Carcinoma Hepatocelular , Hepatite B , Hepatite C , Neoplasias Hepáticas , Humanos , Carcinoma Hepatocelular/diagnóstico , Carcinoma Hepatocelular/etiologia , Carcinoma Hepatocelular/patologia , Neoplasias Hepáticas/diagnóstico , Neoplasias Hepáticas/etiologia , Neoplasias Hepáticas/patologia , Hepatite B/complicações , Hepatite B/diagnóstico , Hepatite C/complicações , Hepatite C/diagnóstico , Vírus da Hepatite B , Hepacivirus , Cirrose Hepática/diagnósticoRESUMO
Introduction: Non-invasive prenatal screening (NIPS) via cell-free DNA (cfDNA) screens for fetal chromosome disorders using maternal plasma, including 22q11.2 deletion syndrome (22q11.2DS). While it is the commonest microdeletion syndrome and has potential implications for perinatal management, prenatal screening for 22q11.2DS carries some inherent technical, biological, and counseling challenges, including varying deletion sizes/locations, maternal 22q11.2 deletions, confirmatory test choice, and variable phenotype. Materials and methods: This study addresses these considerations utilizing a retrospective cohort of 307 samples with screen-positive 22q11.2 NIPS results on a massively parallel sequencing (MPS) platform. Results: Approximately half of the cases reported ultrasound findings at some point during pregnancy. In 63.2% of cases with diagnostic testing, observed positive predictive values were 90.7%-99.4%. cfDNA identified deletions ranging from <1 Mb to 3.55 Mb, with significant differences in confirmed fetal versus maternal deletion sizes; estimated cfDNA deletion size was highly concordant with microarray findings. Mosaicism ratio proved useful in predicting the origin of a deletion (fetal versus maternal). Prediction of deletion size, location, and origin may help guide confirmatory testing. Discussion: The data shows that MPS-based NIPS can screen for 22q11.2DS with a high PPV, and that collaboration between the laboratory and clinicians allows consideration of additional metrics that may guide diagnostic testing and subsequent management.
RESUMO
Timely detection of congenital anomalies using ultrasound improves neonatal care. As specialist sonographers are often geographically dispersed, they are sometimes requested to provide a second opinion via tele-expertise. The present study aimed to evaluate the economic impact of asynchronous tele-expertise in obstetric ultrasound care in private medical practice through a comparison with face-to-face consultations. We conducted a cost minimization analysis using decision tree modeling in order to determine whether asynchronous tele-expertise or face-to-face consultation had the lowest cost, under the assumption of equivalent effectiveness in terms of prenatal diagnosis. Costs were measured from the societal perspective. The data for the base case of our modeling came from a retrospective analysis of the clinical practice of an expert who had been conducting asynchronous tele-expertise for 4 years in France. The study included 260 patients for whom 322 requests for expert opinions were made by physicians/midwives from January 2016 to January 2020. The expected average total cost for tele-expertise for a patient was 74.45 (95% CI: 66.36-82.54) compared to 195.02 (95% CI: 183.90-206.14) for the conventional face-to-face strategy. Accordingly, using tele-expertise led to a statistically significant reduction of 120.57 in the average total cost per patient. A sensitivity analysis confirmed the robustness of the model produced. The results of the present study underline the efficiency of tele-expertise and highlight related economic benefits. Accordingly, they could inform public health policy on the dissemination of tele-expertise in the field of obstetric ultrasound care.
RESUMO
With biotechnological advancements, innovative omics technologies are constantly emerging that have enabled researchers to access multi-layer information from the genome, epigenome, transcriptome, proteome, metabolome, and more. A wealth of omics technologies, including bulk and single-cell omics approaches, have empowered to characterize different molecular layers at unprecedented scale and resolution, providing a holistic view of tumor behavior. Multi-omics analysis allows systematic interrogation of various molecular information at each biological layer while posing tricky challenges regarding how to extract valuable insights from the exponentially increasing amount of multi-omics data. Therefore, efficient algorithms are needed to reduce the dimensionality of the data while simultaneously dissecting the mysteries behind the complex biological processes of cancer. Artificial intelligence has demonstrated the ability to analyze complementary multi-modal data streams within the oncology realm. The coincident development of multi-omics technologies and artificial intelligence algorithms has fuelled the development of cancer precision medicine. Here, we present state-of-the-art omics technologies and outline a roadmap of multi-omics integration analysis using an artificial intelligence strategy. The advances made using artificial intelligence-based multi-omics approaches are described, especially concerning early cancer screening, diagnosis, response assessment, and prognosis prediction. Finally, we discuss the challenges faced in multi-omics analysis, along with tentative future trends in this field. With the increasing application of artificial intelligence in multi-omics analysis, we anticipate a shifting paradigm in precision medicine becoming driven by artificial intelligence-based multi-omics technologies.
Assuntos
Inteligência Artificial , Neoplasias , Humanos , Medicina de Precisão , Multiômica , Neoplasias/diagnóstico , Neoplasias/genética , Neoplasias/terapia , TranscriptomaRESUMO
Metabolomics enables the analysis of metabolites within an organism, which offers the closest direct measurement of the physiological activity of the organism, and has advanced efforts to characterize metabolic states, identify biomarkers, and investigate metabolic pathways. A high degree of innovation in analytical techniques has promoted the application of metabolomics, especially in the study of clinical surgery. Metabolomics can be employed as a clinical testing method to maximize therapeutic outcomes, and has been applied in rapid diagnosis of diseases, timely postoperative monitoring, prognostic assessment, and personalized medicine. This review focuses on the use of mass spectrometry and nuclear magnetic resonance-based metabolomics in clinical surgery, including identifying metabolic changes before and after surgery, finding disease-associated biomarkers, and exploring the potential of personalized therapy. Challenges and opportunities of metabolomics in organ transplantation are also discussed, with a particular emphasis on metabolomics in donor organ evaluation and protection, prognostic outcome prediction, as well as postoperative adverse reaction monitoring. In the end, current limitations of metabolomics in clinical surgery and future research directions are presented.
RESUMO
OBJECTIVES: Health practitioners' understanding of the impact of age-based stereotype threat (ABST) on the cognitive test performance of older adults is not well understood. This study aimed to investigate health practitioners' ability to recognize the influence of ABST in the cognitive assessment of older adults and their perceptions of its impact in practice. METHODS: One-hundred and twenty-nine health practitioners (86% female; M age = 39.75, SD = 11.50) with experience in conducting cognitive assessments with older adults (mainly psychologists and occupational therapists) completed an online survey assessing demographic and practice characteristics, aging beliefs, a hypothetical cognitive assessment scenario, and perceived impact of ABST on practice. RESULTS: Overall, health practitioners rated ABST factors in the assessment scenario as less detrimental to cognitive performance than internal and external factors. In a hierarchical regression model, lower recognition of ABST and negative aging beliefs significantly accounted for lower perceived impact of ABST on older adults' cognitive test performance in practice (R2 = .37, p < .001). CONCLUSION: Health practitioners may not recognize the influence of ABST on assessment findings, especially if they hold negative aging beliefs. The findings highlight the need to improve health practitioners' knowledge of ABST to increase the validity of cognitive testing in older adults.
Assuntos
Envelhecimento , Estereotipagem , Humanos , Feminino , Idoso , Masculino , Envelhecimento/psicologia , Fatores Etários , Inquéritos e Questionários , CogniçãoRESUMO
Spinal muscular atrophy (SMA) is one of the most common genetic disorders worldwide, and genetic testing plays a key role in its diagnosis and prevention. The last decade has seen a continuous flow of new methods for SMA genetic testing that, along with traditional approaches, have affected clinical practice patterns to some degree. Targeting different application scenarios and selecting the appropriate technique for genetic testing have become priorities for optimizing the clinical pathway for SMA. In this review, we summarize the latest technological innovations in genetic testing for SMA, including MassArray®, digital PCR (dPCR), next-generation sequencing (NGS), and third-generation sequencing (TGS). Implementation recommendations for rationally choosing different technical strategies in the tertiary prevention of SMA are also explored.
RESUMO
Aims: Portopulmonary hypertension (PoPH) is a subtype of pulmonary arterial hypertension related to portal hypertension. The definitive diagnosis of PoPH is made by invasive right heart catheterization. Alternatively, pulmonary arterial hypertension may be recognized noninvasively from the tricuspid regurgitant pressure gradient (TRPG), measured by echocardiography. In this study, we aimed to establish a simple algorithm to identify chronic liver disease patients with a high TRPG value in order to narrow down the candidates to receive echocardiography. Methods and Results: TRPG was measured by echocardiography in 152 patients with chronic liver disease. Factors predictive of TRPG >30 mmHg were investigated. There were 28 (18%) cases with TRPG >30 mmHg. Independent factors associated with a high TRPG were the presence of shortness of breath, high serum brain natriuretic peptide (BNP), and low serum albumin. Child-Pugh class or the presence of ascites, varices, or encephalopathy was not associated with TRPG. There was a correlation between the serum BNP and TRPG, and the optimal cutoff value of BNP by the Youden index was 122 pg/mL, and by 100% sensitivity was 50 pg/mL. A combination of these factors identified patients with a high probability of TRPG >30 mmHg (n = 12, positive predictive value [PPV] of 83%), no probability (n = 80, PPV 0%), and intermediate probability (n = 60, PPV 25-34%). This algorithm has reduced the number of patients needing echocardiography by 53%. Conclusions: A simple algorithm using the presence of shortness of breath, serum BNP, and albumin levels can narrow down the candidates to receive echocardiography.
RESUMO
Background and Aim: Nonspecific ileitis is inflammation of the ileum without specific diagnostic features. A minority may go on to develop Crohn's disease, but optimal pathways of further investigation have not been established. This study aimed to identify a cohort of patients with nonspecific ileitis and to determine the value of ileal histology and gastrointestinal ultrasound in identifying/excluding Crohn's disease. Patients and Methods: In a retrospective analysis, all patients having nonspecific ileitis at colonoscopy from January 2010 to August 2021 were identified. Clinical associations with those subsequently diagnosed with Crohn's disease were examined with specific reference to ileal histology and gastrointestinal ultrasound. Results: Of 29 638 procedures, 147 patients (0.5%) had nonspecific ileitis. Crohn's disease was subsequently diagnosed in 8 patients (5.4%) at a median of 148 (range 27-603) days after colonoscopy. The presence of chronic inflammation on ileal biopsies was more common in those subsequently diagnosed with Crohn's disease (63% vs 20%; P = 0.0145). On gastrointestinal ultrasound, none of the 26 patients with normal bowel wall thickness (<3 mm) were subsequently diagnosed with Crohn's disease, and repeat ultrasound in 15 patients 1 year later showed no change. Of the nine patients with abnormal sonographic findings, three were diagnostic for Crohn's disease. Repeat ultrasound revealed Crohn's disease in two, while four had resolution of the abnormal findings. Conclusion: Although ileal histology was of limited value in identifying patients with nonspecific ileitis who were subsequently diagnosed with Crohn's disease, gastrointestinal ultrasound was highly informative. Prospective studies are needed to confirm the value of gastrointestinal ultrasound as a diagnostic and monitoring tool in this setting.
RESUMO
Raman Spectroscopy has long been anticipated to augment clinical decision making, such as classifying oncological samples. Unfortunately, the complexity of Raman data has thus far inhibited their routine use in clinical settings. Traditional machine learning models have been used to help exploit this information, but recent advances in deep learning have the potential to improve the field. However, there are a number of potential pitfalls with both traditional and deep learning models. We conduct a literature review to ascertain the recent machine learning methods used to classify cancers using Raman spectral data. We find that while deep learning models are popular, and ostensibly outperform traditional learning models, there are many methodological considerations which may be leading to an over-estimation of performance; primarily, small sample sizes which compound sub-optimal choices regarding sampling and validation strategies. Amongst several recommendations is a call to collate large benchmark Raman datasets, similar to those that have helped transform digital pathology, which researchers can use to develop and refine deep learning models.
RESUMO
Cervical cancer remains a leading cause of cancer death in women, seriously threatening their physical and mental health. It is an easily preventable cancer with early screening and diagnosis. Although technical advancements have significantly improved the early diagnosis of cervical cancer, accurate diagnosis remains difficult owing to various factors. In recent years, artificial intelligence (AI)-based medical diagnostic applications have been on the rise and have excellent applicability in the screening and diagnosis of cervical cancer. Their benefits include reduced time consumption, reduced need for professional and technical personnel, and no bias owing to subjective factors. We, thus, aimed to discuss how AI can be used in cervical cancer screening and diagnosis, particularly to improve the accuracy of early diagnosis. The application and challenges of using AI in the diagnosis and treatment of cervical cancer are also discussed.
RESUMO
Pancreatic cancer is the second leading cause of cancer-related death in the USA. The 5-year survival rate for pancreatic cancer is as low as 10%, making it one of the most deadly cancers. This dismal prognosis is caused, in part, by the lack of early detection and screening options, leading to late-stage detection of the disease, at a point at which chemotherapy is no longer effective. However, nanoparticle (NP) drug delivery systems have increased the efficacy of chemotherapeutics by improving the targeting ability of drugs to the tumor site, while also decreasing the risk of local and systemic toxicity. Such efforts can contribute to the development of early diagnosis and routine screening tests, which will drastically improve the survival rates and prognosis of patients with pancreatic cancer.
Assuntos
Nanomedicina , Neoplasias Pancreáticas , Detecção Precoce de Câncer , Humanos , Neoplasias Pancreáticas/diagnóstico , Neoplasias Pancreáticas/tratamento farmacológico , Preparações Farmacêuticas , Neoplasias PancreáticasRESUMO
We present the use of conductive spray polymer ionization mass spectrometry (CPSI-MS) combined with machine learning (ML) to rapidly gain the metabolic fingerprint from 1 µl liquid extraction from the biopsied tissue of triple-negative breast cancer (TNBC) in China. The 76 discriminative metabolite markers are verified at the primary carcinoma site and can also be successfully tracked in the serum. The Lasso classifier featured with 15- and 22-metabolites detected by CPSI-MS achieve a sensitivity of 88.8% for rapid serum screening and a specificity of 91.1% for tissue diagnosis, respectively. Finally, the expression levels of their corresponding upstream enzymes and transporters have been initially confirmed. In general, CPSI-MS/ML serves as a cost-effective tool for the rapid screening, diagnosis, and precise characterization for the TNBC metabolism reprogramming in the clinical practice.
RESUMO
Glaucoma is a chronic eye disease, which causes gradual vision loss and eventually blindness. Accurate glaucoma screening at early stage is critical to mitigate its aggravation. Extracting high-quality features are critical in training of classification models. In this paper, we propose a deep ensemble network with attention mechanism that detects glaucoma using optic nerve head stereo images. The network consists of two main sub-components, a deep Convolutional Neural Network that obtains global information and an Attention-Guided Network that localizes optic disc while maintaining beneficial information from other image regions. Both images in a stereo pair are fed into these sub-components, the outputs are fused together to generate the final prediction result. Abundant image features from different views and regions are being extracted, providing compensation when one of the stereo images is of poor quality. The attention-based localization method is trained in a weakly-supervised manner and only image-level annotation is required, which avoids expensive segmentation labelling. Results from real patient images show that our approach increases recall (sensitivity) from the state-of-the-art 88.89% to 95.48%, while maintaining precision and performance stability. The marked reduction in false-negative rate can significantly enhance the chance of successful early diagnosis of glaucoma.