Detection of early lung cancer

Multi-objective deep learning for lung cancer detection in CT images: enhancements in tumor classification, localization, and diagnostic efficiency

OBJECTIVE

This study aims to develop and evaluate an advanced deep learning framework for the detection, classification, and localization of lung tumors in computed tomography (CT) scan images.

MATERIALS AND METHODS

The research utilized a dataset of 1608 CT scan images, including 623 cancerous and 985 non-cancerous cases, all carefully labeled for accurate tumor detection, classification (benign or malignant), and localization. The preprocessing involved optimizing window settings, adjusting slice thickness, and applying advanced data augmentation techniques to enhance the model's robustness and generalizability. The proposed model incorporated innovative components such as transformer-based attention layers, adaptive anchor-free mechanisms, and an improved feature pyramid network. These features enabled the model to efficiently handle detection, classification, and localization tasks. The dataset was split into 70% for training, 15% for validation, and 15% for testing. A multi-task loss function was used to balance the three objectives and optimize the model's performance. Evaluation metrics included mean average precision (mAP), intersection over union (IoU), accuracy, precision, and recall.

RESULTS

The proposed model demonstrated outstanding performance, achieving a mAP of 96.26%, IoU of 95.76%, precision of 98.11%, and recall of 98.83% on the test dataset. It outperformed existing models, including You Only Look Once (YOLO)v9 and YOLOv10, with YOLOv10 achieving a mAP of 95.23% and YOLOv9 achieving 95.70%. The proposed model showed faster convergence, better stability, and superior detection capabilities, particularly in localizing smaller tumors. Its multi-task learning framework significantly improved diagnostic accuracy and operational efficiency.

CONCLUSION

The proposed model offers a robust and scalable solution for lung cancer detection, providing real-time inference, multi-task learning, and high accuracy. It holds significant potential for clinical integration to improve diagnostic outcomes and patient care.

The function of LncRNAs and their role in the prediction, diagnosis, and prognosis of lung cancer

Lung cancer remains a major threat to human health. Low dose CT scan (LDCT) has become the main method of early screening for lung cancer due to the low sensitivity of chest X-ray. However, LDCT not only has a high false positive rate, but also entails risks of overdiagnosis and cumulative radiation exposure. In addition, cumulative radiation by LDCT screening and subsequent follow-up can increase the risk of lung cancer. Many studies have shown that long noncoding RNAs (lncRNAs) remain stable in blood, and profiling of blood has the advantages of being noninvasive, readily accessible and inexpensive. Serum or plasma assay of lncRNAs in blood can be used as a novel detection method to assist LDCT while improving the accuracy of early lung cancer screening. LncRNAs can participate in the regulation of various biological processes. A large number of researches have reported that lncRNAs are key regulators involved in the progression of human cancers through multiple action models. Especially, some lncRNAs can affect various hallmarks of lung cancer. In addition to their diagnostic value, lncRNAs also possess promising potential in other clinical applications toward lung cancer. LncRNAs can be used as predictive markers for chemosensitivity, radiosensitivity, and sensitivity to epidermal growth factor receptor (EGFR)-targeted therapy, and as well markers of prognosis. Different lncRNAs have been implicated to regulate chemosensitivity, radiosensitivity, and sensitivity to EGFR-targeted therapy through diverse mechanisms. Although many challenges need to be addressed in the future, lncRNAs have bright prospects as an adjunct to radiographic methods in the clinical management of lung cancer.

A serum circulating miRNA diagnostic test to identify asymptomatic high-risk individuals with early stage lung cancer

Lung cancer is the first cause of cancer mortality worldwide, and its early detection is currently the main available strategy to improve disease prognosis. While early diagnosis can be successfully achieved through tomography-based population screenings in high-risk individuals, simple methodologies are needed for effective cancer prevention programs. We developed a test, based on the detection of 34 microRNAs (miRNAs) from serum, that could identify patients with early stage non-small cell lung carcinomas (NSCLCs) in a population of asymptomatic high-risk individuals with 80% accuracy. The signature could assign disease probability accurately either in asymptomatic or symptomatic patients, is able to distinguish between benign and malignant lesions, and to capture the onset of the malignant disease in individual patients over time. Thus, our test displays a number of features of clinical relevance that project its utility in programs for the early detection of NSCLC.

Occult lung malignancy presenting with finger pain: a case report

Introduction

Lung cancer is currently one of the most common malignancies in the world. Early detection is an important prognostic factor. Unfortunately, initial symptoms may be vague and a substantial proportion of cases present with the effects of metastases.

Case presentation

We discuss a case of occult lung malignancy in a 61-year-old man. The only symptom at presentation was pain in the right ring finger due to metastasis from the lung primary.

Conclusion

This case highlights the need for vigilance when a patient presents with unusual or unexplained symptoms, especially if they have known risk factors for cancer.

Screening for lung cancer using low dose CT scanning

BACKGROUND

Lung cancer is the most common cause of cancer related death in Ireland. The majority of lung cancers are inoperable at the time of diagnosis and consequently the overall 5 year survival is less than 10%. The objective of the ProActive Lung Cancer Detection (PALCAD) study was to evaluate whether low dose chest computed tomographic scanning (LDCCT) can detect early stage asymptomatic lung cancer in a high risk urban population.

METHODS

Four hundred and forty nine subjects of median age 55 years (range 50-74) with a median pack year smoking history of 45 years (range 10-160), with no previous cancer history and medically fit to undergo thoracic surgery were recruited. After informed consent, LDCCT was performed on all subjects. Non-calcified nodules (NCNs) of >/=10 mm in diameter were referred for biopsy. Follow up with interval LDCCT at 6, 12 and 24 months to exclude growth was recommended for NCNs <10 mm in diameter.

RESULTS

Six (1.3%) NCNs of >/=10 mm were detected of which one (0.23%) had non-small cell lung cancer stage 1; 145 NCNs of <10 mm were detected in 87 (19.4%) subjects. Mediastinal masses were detected in three subjects (0.7%)-one small cell lung cancer and two benign duplication cysts. Incidental pathology was noted in 276 patients (61.5%), most commonly emphysema and coronary artery calcification.

CONCLUSION

The prevalence of resectable lung cancer detected by LDCCT at baseline screening was low at 0.23%, but there was a high rate of significant incidental pathology.

The pitfalls of lung cancer screening

Lung cancer screening has received extensive attention for a number of years. As yet the goal of such a screening programme, a reduction in lung cancer mortality proven by a large randomised controlled trial, has not been achieved. Instead we are left with a number of unanswered questions and practical problems. In addition to the basic requirements for an effective screening programme, this review will identify the main pitfalls in lung cancer screening, with particular reference to multislice computed tomography. The specific difficulties relating to the identification of unimportant disease, the failure to identify important disease successfully, the consequences of investigating and treating identified disease and the financial costs will all be discussed.

[Delays in the diagnosis of lung cancer]

OBJECTIVE

To study the clinical and demographic factors associated with delays in the diagnosis of lung cancer.

PATIENTS AND METHODS

A 2-year prospective study of patients admitted to the respiratory medicine ward with a suspected diagnosis of lung cancer. We studied demographic factors, health care received, place of residence, and delays in carrying out diagnostic procedures. The following diagnostic time periods were defined: consultation (from first symptom to first medical visit), middle period (from first medical visit to hospital admission) and diagnostic (from hospital admission to histological diagnosis and clinical staging).

RESULTS

One hundred thirteen patients with a mean age of 65 years (range, 36-90), 103 men and 10 women, were studied. The most frequent symptoms leading to consultation were coughing (10.6%), hemoptysis (19.5%), chest pain (26.5%), and shortness of breath (9.7%). First visits were to a primary care physician for 72%, to the hospital emergency room for 22%, or to a pulmonologist for 6%. Forty-four percent of the patients visited the doctor 2 or 3 times. The mean SD, numbers of days for the different time periods were as follows: consultation, 30.3 60; diagnosis, 18.6 19; middle period 37.9 63. The mean total time from first symptom to diagnosis was 85.7 87 days. The middle period, the time in hospital until diagnosis, and the total time were shorter when patients were referred by the primary care physician to the emergency room or were directly admitted to the hospital (P<.001). Only 25.7% of the staged lung cancers were operable.

CONCLUSIONS

Delays in lung cancer diagnosis are long. The attitudes of primary care physicians and their relations with specialized care providers are crucial for reducing delays.

Screening for early lung cancer with low-dose spiral CT: prevalence in 817 asymptomatic smokers

PURPOSE

To present prevalence screening data from a nonrandomized screening trial by using low-dose computed tomography (CT) and a simple algorithm based on the size and attenuation of detected nodules to guide diagnostic work-up.

MATERIALS AND METHODS

Eight hundred seventeen asymptomatic volunteers (age range, 40-78 years; median age, 53 years; median tobacco consumption, 45 pack-years) underwent spiral low-dose CT of the chest without contrast material enhancement. We regarded all noncalcified pulmonary nodules greater than 10 mm in diameter as potentially malignant and recommended histologic examination or follow-up after 3, 6, 12, and 24 months to exclude growth. For noncalcified pulmonary nodules of 10 mm or smaller, repeat low-dose CT was recommended to exclude growth.

RESULTS

In 43% (350 of 817) of individuals, 858 noncalcified pulmonary nodules were found. Thirty-two nodules in 29 subjects were larger than 10 mm. Biopsy of 15 lesions revealed lung cancer in 12 lesions in 11 subjects (prevalence for all ages, 1.3% [11 of 817 subjects]; >50 years of age, 2.1% [11 of 519 subjects]; >60 years of age, 3.9% [eight of 206 subjects]), with a high proportion of early tumor stages (seven tumors, stage I; two, stage II; and three, stage III); three lesions were benign. In 17 nodules larger than 10 mm, follow-up with low-dose CT for a minimum of 24 months did not demonstrate growth.

CONCLUSION

Lung cancer screening with low-dose CT demonstrated a prevalence of asymptomatic cancers in 1.3% of a smoking population, including a high proportion of early tumor stages and a 20% (three of 15) rate of invasive procedures for benign lesions.

Lung cancer screening

Prior attempts to screen for lung cancer using chest radiography and sputum cytology have proved unsuccessful. Recent reports have investigated the role of spiral CT in early lung cancer detection and have suggested that screening would be of value. Prior to the introduction of a national lung cancer screening programme, it would be essential to demonstrate that this would reduce mortality and would be cost effective.