AI-driven Characterization of Solid Pulmonary Nodules on CT Imaging for Enhanced Malignancy Prediction in Small-sized Lung Adenocarcinoma.

OBJECTIVES: Distinguishing solid nodules from nodules with ground-glass lesions in lung cancer is a critical diagnostic challenge, especially for tumors ≤2 cm. Human assessment of these nodules is associated with high inter-observer variability, which is why an objective and reliable diagnostic tool is necessary. This study focuses on artificial intelligence (AI) to automatically analyze such tumors and to develop prospective AI systems that can independently differentiate highly malignant nodules. MATERIALS AND METHODS: Our retrospective study analyzed 246 patients who were diagnosed with negative clinical lymph node metastases (cN0) using positron emission tomography-computed tomography (PET/CT) imaging and underwent surgical resection for lung adenocarcinoma. AI detected tumor sizes ≤2 cm in these patients. By utilizing AI to classify these nodules as solid (AI_solid) or non-solid (non-AI_solid) based on confidence scores, we aim to correlate AI determinations with pathological findings, thereby advancing the precision of preoperative assessments. RESULTS: Solid nodules identified by AI with a confidence score ≥0.87 showed significantly higher solid component volumes and proportions in patients with AI_solid than in those with non-AI_solid, with no differences in overall diameter or total volume of the tumors. Among patients with AI_solid, 16% demonstrated lymph node metastasis, and a significant 94% harbored invasive adenocarcinoma. Additionally, 44% were upstaging postoperatively. These AI_solid nodules represented high-grade malignancies. CONCLUSION: In small-sized lung cancer diagnosed as cN0, AI automatically identifies tumors as solid nodules ≤2 cm and evaluates their malignancy preoperatively. The AI classification can inform lymph node assessment necessity in sublobar resections, reflecting metastatic potential.

Development of automatic generation system for lung nodule finding descriptions.

Worldwide, lung cancer is the leading cause of cancer-related deaths. To manage lung nodules, radiologists observe computed tomography images, review various imaging findings, and record these in radiology reports. The report contents should be of high quality and uniform regardless of the radiologist. Here, we propose an artificial intelligence system that automatically generates descriptions related to lung nodules in computed tomography images. Our system consists of an image recognition method for extracting contents-namely, bronchopulmonary segments and nodule characteristics from images-and a natural language processing method to generate fluent descriptions. To verify our system's clinical usefulness, we conducted an experiment in which two radiologists created nodule descriptions of findings using our system. Through our system, the similarity of the described contents between the two radiologists (p = 0.001) and the comprehensiveness of the contents (p = 0.025) improved, while the accuracy did not significantly deteriorate (p = 0.484).

Development and performance evaluation of a deep learning lung nodule detection system.

BACKGROUND: Lung cancer is the leading cause of cancer-related deaths throughout the world. Chest computed tomography (CT) is now widely used in the screening and diagnosis of lung cancer due to its effectiveness. Radiologists must identify each small nodule shadow from 3D volume images, which is very burdensome and often results in missed nodules. To address these challenges, we developed a computer-aided detection (CAD) system that automatically detects lung nodules in CT images. METHODS: A total of 1997 chest CT scans were collected for algorithm development. The algorithm was designed using deep learning technology. In addition to evaluating detection performance on various public datasets, its robustness to changes in radiation dose was assessed by a phantom study. To investigate the clinical usefulness of the CAD system, a reader study was conducted with 10 doctors, including inexperienced and expert readers. This study investigated whether the use of the CAD as a second reader could prevent nodular lesions in lungs that require follow-up examinations from being overlooked. Analysis was performed using the Jackknife Free-Response Receiver-Operating Characteristic (JAFROC). RESULTS: The CAD system achieved sensitivity of 0.98/0.96 at 3.1/7.25 false positives per case on two public datasets. Sensitivity did not change within the range of practical doses for a study using a phantom. A second reader study showed that the use of this system significantly improved the detection ability of nodules that could be picked up clinically (p = 0.026). CONCLUSIONS: We developed a deep learning-based CAD system that is robust to imaging conditions. Using this system as a second reader increased detection performance.

Direct evidence of EEG coherence in alleviating phantom limb pain by virtual referred sensation: Case report.

Virtual reality (VR) systems have been integrated into rehabilitation techniques for phantom limb pain (PLP). In this case report, we used electroencephalography (EEG) to analyze corticocortical coherence between the bilateral sensorimotor cortices during vibrotactile stimulation in conjunction with VR rehabilitation in two PLP patients. As a result, we observed PLP alleviation and increased alpha wave coherence during VR rehabilitation when stimulation was delivered to the cheek and shoulder (referred sensation areas) of the affected side. Vibrotactile stimulation with VR rehabilitation may enhance the awareness and movement of the phantom hand.

Somatosensory Feedback to the Cheek During Virtual Visual Feedback Therapy Enhances Pain Alleviation for Phantom Arms.

BACKGROUND: Patients who suffer from phantom limb pain can perceive tactile stimuli applied to the cheek on their affected side as if it were coming from their phantom limb, a phenomenon called "referred sensation." OBJECTIVES: To investigate the analgesic effect produced by tactile feedback provided to the cheek during neurorehabilitation using visual feedback. METHODS: Nine participants with phantom upper limb pain performed virtual reality neurorehabilitation exercises in which they repeatedly touched a target object with a virtual representation of their affected limb. We applied tactile feedback to their cheek when their virtual affected limb touched a virtual object (Cheek Condition). We also included 2 control conditions where tactile feedback was either applied to their intact hand (Intact Hand Condition) or not applied at all (No Stimulus Condition). We evaluated pain intensity on an 11-point rating scale and pain quality using the short-form McGill Pain Questionnaire before and after each rehabilitation condition. RESULTS: The median pain-reduction rate in the Cheek Condition (33.3 ± 24.4%) was significantly higher than in the Intact Hand Condition (16.7 ± 12.3%) and the No Stimulus Condition (12.5 ± 13.5%; P < .05). Even patients who did not feel referred sensations reported significant pain reduction after the Cheek Condition. CONCLUSIONS: The analgesic effect of neurorehabilitative visual feedback during phantom limb movement is significantly improved by applying somatosensory feedback to the cheek on the affected side. Further studies are needed to extend these findings to objective pain measures and to elucidate the neural mechanisms that underlie the analgesic effect.

Tactile feedback for relief of deafferentation pain using virtual reality system: a pilot study.

BACKGROUND: Previous studies have tried to relieve deafferentation pain (DP) by using virtual reality rehabilitation systems. However, the effectiveness of multimodal sensory feedback was not validated. The objective of this study is to relieve DP by neurorehabilitation using a virtual reality system with multimodal sensory feedback and to validate the efficacy of tactile feedback on immediate pain reduction. METHODS: We have developed a virtual reality rehabilitation system with multimodal sensory feedback and applied it to seven patients with DP caused by brachial plexus avulsion or arm amputation. The patients executed a reaching task using the virtual phantom limb manipulated by their real intact limb. The reaching task was conducted under two conditions: one with tactile feedback on the intact hand and one without. The pain intensity was evaluated through a questionnaire. RESULTS: We found that the task with the tactile feedback reduced DP more (41.8 ± 19.8 %) than the task without the tactile feedback (28.2 ± 29.5 %), which was supported by a Wilcoxon signed-rank test result (p < 0.05). CONCLUSIONS: Overall, our findings indicate that the tactile feedback improves the immediate pain intensity through rehabilitation using our virtual reality system.

Structured movement representations of a phantom limb associated with phantom limb pain.

The relation between phantom limb pain (PLP) and the movement representation of a phantom limb remains controversial in several areas of neurorehabilitation, although there are a few studies in which the representation of phantom limb movement was precisely evaluated. We evaluated the structured movement representation of a phantom limb objectively using a bimanual circle-line coordination task. We then investigated the relation between PLP and the structured movement representation. Nine patients with a brachial plexus avulsion injury were enrolled who perceived a phantom limb and had neuropathic pain. While blindfolded, the participants repeatedly drew vertical lines using the intact hand and intended to draw circles using the phantom limb simultaneously. "Drawing of circles" by the phantom limb resulted in an oval transfiguration of the vertical lines ("bimanual coupling" effect). We used an arbitrary ovalization index (OI) to quantify the oval transfiguration. When the OI neared 100%, the trajectory changed toward becoming more circular. A significant negative correlation was observed between the intensity of PLP and the OI (r=-0.66, p<0.05). Our findings directly suggest that structured movement representations of the phantom limb are necessary for alleviating PLP.

Reliability of phantom pain relief in neurorehabilitation using a multimodal virtual reality system.

The objective of this study is to demonstrate the reliability of relief from phantom limb pain in neurore-habilitation using a multimodal virtual reality system. We have developed a virtual reality rehabilitation system with multimodal sensory feedback and applied it to six patients with brachial plexus avulsion or arm amputation. In an experiment, patients executed a reaching task using a virtual phantom limb displayed in a three-dimensional computer graphic environment manipulated by their real intact limb. The intensity of the phantom limb pain was evaluated through a short-form McGill pain questionnaire. The experiments were conducted twice on different days at more than four-week intervals for each patient. The reliability of our task's ability to relieve pain was demonstrated by the test-retest method, which checks the degree of the relative similarity between the pain reduction rates in two experiments using Fisher's intraclass correlation coefficient (ICC). The ICC was 0.737, indicating sufficient reproducibility of our task. The average of the reduction rates across participants was 50.2%, and it was significantly different from 0 (p <; 0:001). Overall, our findings indicate that neurorehabilitation using our multimodal virtual reality system reduces the phantom limb pain with sufficient reliability.