‘Feeling Hot’: Exploring the feasibility of nocturnal erection detection through penile temperature measurements

Abstract Objectives The observational ‘Feeling Hot’ study aims to evaluate the feasibility of employing overnight penile temperature measurements for the detection of nocturnal erections, thereby contributing to the advancement and modernization of a non‐invasive diagnostic system for erectile dysfunction. Subjects/Patients and Methods In this proof‐of‐concept study, 10 healthy men aged 20–25 were recruited, following the methodology outlined in the ‘Staying Hot’ study by Torenvlied et al. Participants underwent ambulatory overnight penile temperature measurements concurrent with RigiScan recordings. Key outcome measures included baseline and peak penile temperatures during RigiScan‐annotated nocturnal erections. Reference measurements of the thigh temperature were also taken to assess nocturnal temperature variations. Results Statistically significant penile temperature increases (p = 0.008, n = 9) were observed during nocturnal erections, with an average elevation of 1.47°C noted during the initial erections. This underscores the practical utility of penile temperature measurements in detecting erection onset. Challenges arose in accurately determining erection duration and subsequent erection onsets due to the persistence of elevated temperatures following initial erections, termed the ‘Staying Hot effect’. Reference thigh temperature measurements aided in addressing this challenge. Conclusion Examining overnight penile temperature alongside simultaneous RigiScan recordings has yielded valuable insights into the viability of using the temperature methodology for detecting nocturnal erections. The ‘Feeling Hot’ study findings demonstrate significant penile temperature elevation during nocturnal erections in healthy young men, highlighting the potential of integrating this measurement methodology into the design of a modernized tool for ambulatory erectile dysfunction diagnostics. Further development of an advanced sensor system to comprehensively assess erection duration and quality is essential for enhancing clinical applicability.


| INTRODUCTION
Erectile dysfunction (ED), defined as 'the inability to attain and maintain an erection sufficient to permit satisfactory sexual performance', is a common benign condition affecting up to 52% of men aged 40-70 years. 1 Despite its benign nature, ED significantly impacts the quality of life for both patients and their families. 2Accurately distinguishing between organic and psychogenic ED is crucial for effective treatment.Currently, the RigiScan stands as the gold standard diagnostic tool for detecting nocturnal penile erections. 3,4Nocturnal erections, occurring automatically and involuntarily during rapid eye movement (REM) sleep, serve as a reliable indicator for normal erectile function. 5,6However, the clinical utility of the RigiScan is diminishing due to its cumbersome size, outdated software and suboptimal patient experience, raising concerns about its validity. 4,7Consequently, many healthcare institutions resort to invasive alternative diagnostic techniques, emphasizing the urgent need for a modernized system for diagnosing ED.
Non-invasive measurement of penile rigidity and circumference relies on incorporating a pressure component within the measurement system, which can impact patients' sleep quality and consequently the validity of overnight measurements.10][11][12] This raises the question of whether detecting nocturnal erections through penile temperature measurements, which offer a nonpressure approach, is feasible.The 'Feeling Hot' study aims to address this inquiry by examining penile temperature measurements during RigiScan-annotated nocturnal erections.Accordingly, we recruited 10 healthy participants aged 20-25, strictly adhering to the exclusion criteria specified in the referenced research.In summary, exclusion criteria included unwillingness to provide informed consent, an International Index of Erectile Function (IIEF)-5 score below 22 and a medical history involving sickle cell anaemia, atherosclerosis, diabetes or the use of benzodiazepines. 13Furthermore, the exclusion criteria for the 'Feeling Hot' study were expanded to encompass a medical history of REM-sleep behaviour disorder, restless legs syndrome, insomnia, sleep apnoea and the use of sleeping pills.

| Study setting and participants
Participants were included between April and June 2022.

| Materials and study procedure
The temperature measurement system utilized in the 'Feeling Hot' study was an expanded version of the temperature sensing system employed in the 'Staying Hot' study. 12The system comprised three Ohmeda 3P T3312 temperature probes (Shenzhen Medke Technology Co., Shenzhen, Guangdong, China), 14 connected to a PicoLog 1216 data logger (Pico Technology, Cambridgeshire, UK), 15 and a standalone Lenovo ThinkBook 13s-IWL laptop.Positioning the system near the participants' bed facilitated unrestricted movement and reduced entanglement hazards.
Participants were instructed to sleep in cotton underwear and under a blanket, alone and refrain from alcohol consumption, sexual activity or masturbation on the evening of participation.To ensure participant privacy and minimize skin contact duration, participants themselves performed the sensor placement.
The first thermistor temperature sensor was positioned on the dorsal penile skin proximal to the glans.The second temperature sensor, the reference sensor, was placed on the subject's thigh.Both sensors were secured to the skin using Leukosan 12 Â 100-mm strips.The third temperature sensor, placed near the participant's bed, recorded room temperature, ensuring consistency across the study cohort.
Nocturnal erections were annotated using the RigiScan (GOTOP Medical Inc., St. Paul, MN, USA). 3 Participants followed the standard protocol, placing the two RigiScan loops at the tip and base of the penis and positioning the datalogger in the designated fabric on the subjects' legs.Following sensor placement, participants proceeded to take overnight ambulatory measurements upon entering the bed.
The success criteria for the overnight measurements involved the detection of at least one nocturnal erection lasting a minimum of 15 min, as recorded by the RigiScan device. 3In the event of missing data, test subjects were excluded.

| Data processing and statistical analysis
The temperature data were processed using MATLAB R2019a (The MathWorks, USA), employing the methodologies outlined in the 'Staying Hot' study by Torenvlied et al. 12 RigiScan images were imported into MATLAB, and the 'ginput' tool was utilized to extract the precise start and endpoint of nocturnal erections following the standard protocol. 3Four outcome variables were defined for temperature data during the annotated erectile periods: • T peak : the peak temperature during the erectile period; • T baseline : the minimum temperature in the 15 min preceding the RigiScan-annotated erectile period; • ΔT: the temperature difference between T peak and T baseline (ΔT ¼ T peak À T baseline Þ; and • erection duration. To address changes in body temperature resulting from sleeping under blankets and to mitigate temperature fluctuations due to movement, the reference sensor was utilized.The sensor's data were analysed to derive the 'corrected penile temperature', defined as the disparity between penile and thigh temperature (T penis À T thigh ).Values for T peak , T baseline and ΔT were computed for both the penile temperature data and the 'corrected penile temperature'.
In addition to calculating the aforementioned variables, a visual assessment of the temperature data was conducted.This involved analysing the curves of both the raw penile temperature and the corrected penile temperature data.Furthermore, slope analysis was performed by determining the derivative (dT/dt) of the temperature curve with a time interval (dt) set at 20 s.To mitigate short-term disturbances in penile temperature, temperature values within this 20-s timeframe were averaged.
The statistical analysis was carried out using IBM SPSS Statistics (Version 25.0, Armonk, NY, USA).Initially, the data underwent normal distribution testing through the Shapiro-Wilk test.Subsequently, the Wilcoxon signed-rank test was employed to assess significant differences between baseline and peak temperature (ΔT).Correlations were computed using Spearman's rho.The threshold for statistical significance was set at α < 0.05.

| RESULTS
In this proof-of-concept study, a total of 10 overnight measurements Table 1 illustrates penile temperature variables for overnight erectile periods.The difference between penile baseline and peak temperature was significant ( p < 0.01), with an average ΔT of 0.87 C (SD 0.31).The initial nocturnal erection displayed the highest ΔT, with an average increase of 1.47 C (SD 0.87).Figure 1 visually represents the penile temperature increase during the initial erection of all participants.The ΔT for the initial erection was significantly higher than subsequent erections due to the significantly lower baseline temperature.
The corrected temperature (T penis À T thigh ) data mirrored penile temperature results as shown in Table 2.A significant difference existed between baseline and peak temperatures (p = 0.008).
Figure 2 provides a comprehensive visual of (corrected) penile temperature, slope analysis, and simultaneously measured RigiScan outcomes for a single participant.The initial 15 min demonstrated a T A B L E 1 Overview of the outcomes of the penile temperature measurements for the naked and blanket measurements (n = 9).Values are given as the mean (SD) or median (IQR), dependent on the distribution of the data.T baseline is the minimal temperature in the 15 min prior to erection, T peak is the maximal temperature during erection and ΔT is the difference between T baseline and T peak during erection.Preceding the 'Feeling Hot' study was the 'Staying Hot' study, which explored the impact of overnight conditions on erectile penile temperature in a controlled environment. 12The study found an average penile temperature increase of 0.67 C during erection measurements, prompting the question of whether this increase would be adequate for detecting nocturnal erections.[18] Despite the notable difference in ΔT observed between the two studies, the 'Staying Hot effect' distinctly emerges in the overnight penile temperature curves.This effect implies that overnight conditions, such as the presence of blankets and underwear, lead to the Outcomes of a single overnight measurement of the penile temperature, corrected temperature, penile temperature slope analysis and RigiScan rigidity and circumference measurements.The start and end of the nocturnal erectile phases, according to the RigiScan circumference data, are indicated with green and red lines, respectively.
prolonged maintenance of penile temperature at its peak for over 30 min following the conclusion of an erection. 12In the overnight measurements, it was observed that the penile temperature did not revert to baseline values after the initial erection, substantiated by the significant difference in baseline temperature between the initial and subsequent erections.
The influence of the 'Staying Hot effect' on the reliability of nocturnal erection detection using the temperature methodology sparks a discussion, particularly in the challenging detection of subsequent nocturnal erections.However, it is crucial to note that the disclosure of the presence of a single erectile period is adequate for diagnosing psychogenic ED, aligning with the primary objective of conducting ambulatory measurements.The initial erection was conspicuously distinguishable in the penile temperature curves of all participants during RigiScan-annotated nocturnal erections.Consequently, nocturnal erection detection through temperature measurement not only proves feasible but also exhibits diagnostic potential for distinguishing the nature of ED.
The viability of detecting the presence of nocturnal erections using the temperature methodology is evident; however, improving the capability for subsequent erection detection would be beneficial.While the temperature sensor does not exert pressure on the penile tissue, the results of the 'Feeling Hot' study reveal a limitation in the methodology-it only allows for annotating the initiation of an erection.Notably, it cannot discern the commencement of detumescence, making the determination of erection duration impossible.In contrast, the RigiScan not only provides insights into this duration but also assesses erectile quality through penile rigidity measurements.An ideal modernized diagnostic tool should encompass non-pressure measurement of both erection duration and erectile quality.This could be achieved by integrating multiple physiological principles into a unified (wireless) sensor system, considering factors like penile saturation, pulse or acceleration, especially in line with the goal of avoiding pressure on the penile tissue. 19e development of such a sensor system lays the foundation for clinical implementation, which would naturally follow the execution of a validity study conducted in a larger study population.The feasibility of erection detection with the temperature methodology was proven for healthy young men in this study.For validity studies, the system's capability to differentiate between somatic and psychogenic causes in ED patients should be investigated.Understanding the impact of aging or the absence of erections on overnight (temperature) curves is crucial.Additionally, insights into the demographic properties of the ED patient population are necessary to ensure a comprehensive design of future validity and implementation studies.
The observational 'Feeling Hot' study, conducted at the Department of Urology, St. Antonius Ziekenhuis, Nieuwegein, the Netherlands, received ethical approval from the medical research ethics committees united on 23 February 2022 (NL79969.100.21,R21.115).Prior to participant enrolment, the proof-of-concept study was pre-registered at ClinicalTrials.gov(NCT05183581).The findings of the 'Staying Hot' study indicate an expected increase in penile temperature during nocturnal erections of 0.67 C (SD 0.41).A power calculation using repeated measures with α < 0.05 and β = 20% determined a required sample size of n = 5.The inclusion criteria of the 'Feeling Hot' study closely align with those outlined in the 'Staying Hot' study by Torenvlied et al.
were performed.One participant was excluded due to sleep disturbances caused by the RigiScan, preventing sleep.The mean age of participants was 22.1 years (SD 1.97), with a body mass index (BMI) of 22.00 kg/m 2 and a median IIEF-5 score of 24.00 (interquartile range [IQR] 2.00).Among the remaining nine participants, the RigiScan detected a total of 32 erectile periods, averaging 3.56 periods per subject (SD 1.13) with an average duration of 36.67 min (SD 11.34).The room temperature during the ambulatory measurement averaged 21.79 C (SD 2.69), showing stable overnight curves for each participant.No significant correlation was found between room temperature and baseline temperature (Spearman's rho = 0.633, p > 0.05).

Figure 2
Figure 2 also shows that following the initial nocturnal erection and during subsequent nocturnal erections, the penile temperature remained at an elevated peak level.The corrected temperature course and slope analysis do correspond to the RigiScan-annotated rigidity measurements.Clear increases in temperature were observed during the second and third erectile phases in the curve of corrected temperature, indicating an increase in penile temperature relative to the environmental conditions.A slope analysis of the penile temperature provided a more precise measurement of the start and end times of the nocturnal erections, with clear fluctuations in the slope occurring during all three erections.These fluctuations were absent during the non-erectile phases.Most overnight curves exhibited patterns similar to Figure 2, with clear detection of an initial penile temperature increase for all

Figure 3
Figure 3 illustrates overnight data for one participant with undetectable subsequent nocturnal erections.The corrected temperature showed a decrease during the erections and fluctuations in slope analysis did not align with RigiScan-annotated rigidity measurements.Consequently, visual detection of subsequent nocturnal erections was not possible for this participant.Additionally, Figure 3 highlights a distinct phenomenon observed in all test subjects within the initial 15 min of overnight measurements.Following the onset of the temperature sensors and RigiScan, the penile temperature exhibited an approximate increase of 2 C.However, the corrected temperature also displayed an increase of approximately 1 C instead of stabilizing or decreasing.This indicates that the penile temperature experienced a greater increase than what could be attributed to environmental conditions alone.Analysis of the RigiScan data reveals the presence of small rigidity peaks and fluctuations in circumference, suggesting a sexual stimulus.This phenomenon was observed in three subjects.

1 .
47 C during the initial erection, affirming the feasibility of detecting nocturnal erections in overnight measurements.Two factors explain the difference in temperature change between the studies.First, the average duration of nocturnal erections in this study was 36.67 min, compared with 7.47 min in the 'Staying Hot' study.Second, the baseline temperature was 31.73C, as opposed to 32.75 C. While the human body temperature remains stable or rises during sexual arousal,

Figure 2
Figure 2 illustrates that the curves of corrected temperature and slope analysis enabled the visual identification of subsequent nocturnal erections, a capability observed in seven out of the nine participants.It is highly probable that the application of advanced data analysis techniques, such as machine learning techniques, could facilitate the identification of subsequent nocturnal erections in temperature data.Further research is necessary to explore the feasibility and effectiveness of implementing these techniques.

F I G U R E 3
Visualization of the added value of corrected temperature measurements and slope analysis on the detection of (subsequent) nocturnal erections.An overview is given of the penile temperature, corrected temperature, penile temperature slope analysis and RigiScan rigidity and circumference measurements.The start and end of the nocturnal erectile phases, according to the RigiScan circumference data, are indicated with green and red lines, respectively.

5 |
CONCLUSION Examining overnight penile temperature alongside simultaneous RigiScan recordings has provided valuable insights into the potential of utilizing temperature-based methods for detecting nocturnal erections.The findings of the 'Feeling Hot' study reveal a significant penile temperature increase during nocturnal erections in healthy young men, underscoring the feasibility of integrating this approach into a modernized ambulatory diagnostic tool for ED.Nevertheless, further refinement and extension of an advanced sensor system are essential to comprehensively evaluating erection duration and quality, addressing challenges posed to the temperature by the 'Staying Hot effect'.Despite these challenges, the 'Feeling Hot' study represents a significant initial advancement towards modernizing non-invasive ED diagnostics, paving the way for the reintroduction of patientfriendly ED diagnostic solutions in clinical settings worldwide.