Oncologists nowadays are faced with big amount of heterogeneous medical data of diagnostic studies. Possible errors in determining the nature and extent of spread the tumor process will inevitably reduce the effectiveness of treatment and increase the unnecessary costs to it. To reduce the burden on clinicians, various computer-aided solutions based on machine learning algorithms are being developed. We made an attempt to evaluate effectiveness of thirteen machine learning algorithms in the tasks of classification of pathologic tissue samples in cancerous thorax based on gene expression levels. For a preliminary study we used open data set of molecular genetics composition of lung adenocarcinoma and pleural mesothelioma. Effectiveness of machine learning algorithms was evaluated by Matthews correlation coefficient and Area Under ROC Curve. Best results were showed by two methods: Bayesian logistic regression and Discriminative Multinomial Naive Bayes classifier. Nevertheless, all of the methods were effective at automatic discrimination of two types of cancer. That proves machine learning algorithms are applicable in lung cancer classification. In the future studies it will be carried out a similar analysis of the diagnostic value of methods for other malignancies with more complex differential morphological diagnosis. Similar methods can be applied to other diagnostic studies including computerized tomography image analysis in the differential diagnosis of lung nodules.
Опыт работы60 В в е д е н и е. Заболеваемость раком лёгкого (РЛ) в мире достигает 1 300 000 случаев [3,6,8,12,16]. В Российской Федерации РЛ находится на 2-м месте в общей структуре онкологических заболеваний (11,6 %) и на 1-м (20,4 %) среди злокачественных опухолей у мужчин. Число заболевших им ежегодно превышает 50 000 человек, причём умирают от него 90-96 % заболевших, а свыше 50 % из нихуже в год постановки диагноза. Несмотря на бурное развитие диагностических и лечебных технологий, 5-летняя выживаемость при РЛ во всех странах за последние полвека не меняется, не превышая 15-20 %. При этом известно, что эффективность лечения находится в прямой зависимости от распространённости опухоли на момент начала реализации клинических мероприятий. Так, если для I стадии 5-летняя выживаемость может достигать 70-80 %, то для IV -не превышает 5 %. Ранняя диагностика РЛ до настоящего времени остаётся нерешённой проблемой, и более 2 / 3 заболевших начинают специализированное лечение, имея местно-распространённые либо генерализованные формы опухоли [1, 6, 7, 13]. когортное исследоВАние эффектиВности низкодозной коМПьютерной тоМогрАфии и трАнсторАкАльной треПАн-биоПсии В рАнней диАгностике рАкА лёгкогоЦеЛь иССЛедОВАНия. Анализ литературных и собственных данных для повышения эффективности выявления рака л¸гкого (РЛ) с использованием современных методов первичной и уточняющей диагностики. МАтеРиАЛ и МетОды. Проспективные диагностические данные в отношении контингента из 537 человек, состоящего из двух групп: 1) когортных исследований по ранней диагностике РЛ с использованием низкодозной компьютерной томографии (n=369) и 2) трансторакальной трепан-биопсии (n=168). РеЗуЛьтАты. Патологические изменения в паренхиме л¸гких, подозрительные на ранний периферический рак, выявлены у 24 % участников исследования. Эффективность использования в качестве метода уточняющей диагностики трансторакальной трепан-биопсии составила 85,7 %, находясь в прямой зависимости от размеров очагов и их расположения в паренхиме л¸гких. ЗАКЛючеНие. Современные методы диагностики РЛ позволяют обнаружить заболевание на ранних стадиях и получить достаточные по объ¸му образцы патологической ткани с целью индивидуализации алгоритмов лечения. Ключевые слова: рак л¸гкого, низкодозная компьютерная томография, скрининг, трансторакальная стереотаксическая трепан-биопсия Panel study of the effectiveness of low-dose computed tomography and transthoracic core biopsy in early diagnostics of lung cancer 1 st. Petersburg clinical Research center of specialized types of healthcare (oncological); 2 Petrov Research Institute of oncology, Ministry of healthcare of the Russian federation, saint-Petersburg objecTIve. This paper reviews literature and provides results of original trial data on early diagnostics of lung cancer (lc) with primary and work-up diagnostic procedures. MATeRIAl AnD MeThoDs. The pospective diagnostic data of 537 patients divided into 2 groups was analyzed: 1) panel study in early diagnostics of lc using low-dose computed tomography (369 patients) and 2) tran...
Cancer screening literature was discussed in this review publication. Broad spectrum of studies was used to make conclusion about effectiveness of screening methods in reaching its major objectives, perspective of screening methods for several cancer types were also discussed. Qualitative assessment of studies was done. Cervical cancer, breast cancer and colorectal cancer screening was proved to be effective. Effectiveness of prostate and lung cancer screening as well as population-based stomach cancer prevention is also discussed. Negative and inconclusive results of screening studies of the other cancer types were also mentioned and perspectives for future diagnostics option for cancer screening were given.
This article reviews the literature and summarizes single institution experience of applying different diagnostic algorithms for lung cancer. All diagnostic methods can be divided into three groups: non-invasive; minimally invasive and invasive. The non-invasive methods include clinical examination; imaging methods for anatomical, functional and multimodal visualization; sputum cytological, analysis of the exhaled breath, detection of various blood and sputum markers. Minimally invasive methods include endoscopy, percutaneous fine-needle and core-needle biopsy. Invasive methods include diagnostic thoracoscopy and laparoscopy, mediastinoscopy, parasternal mediastinotomy and diagnostic thoracotomy. While creating an individual diagnostic plan for each patient it is necessary to carefully analyze the effectiveness, safety, sensitivity, specificity and of different methods available among wide range of modern diagnostic techniques. Optimization of lung cancer diagnosis methods, which includes early cancer detection, is one of priority areas of modern oncology. Many aspects of this problem remain unresolved and require further research
In this article we summarize our own experience of lung cancer diagnostics using exhaled breath analysis with a non-selective method using metal oxide chemoresistor gas sensors with cross-sensitivity combined with the sputum cytology. Volatile organic compounds of exhaled breath change the conductivity of the sensor, the resulting pulse is displayed as a peak on the graph, the area of which is used as test results. The combination of two diagnostic techniques in 204 participants demonstrated the possibility of non-invasively detecting the disease at an early stage. The sensitivity, specificity and accuracy of the breath analysis was 91.2%, 100% and 93.4%, respectively. The combination of the breath test and the sputum cytology compared to the breath test alone showed statistically significant (p = 0.03) increase in sensitivity to 96.8% (95% CI: 80.9% -99%) with acceptable decrease in specificity to 93.4% (95% CI: 88% -96%). The convenience of analysis and realtime measurements show some promise for the early detection.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.